In this notebook we will train models on the mobility data.
import os
import pandas as pd
from google.colab import auth
from datetime import datetime
auth.authenticate_user()
!wandb login
!gcloud source repos clone github_aistream-peelout_flow-forecast --project=gmap-997
os.chdir('/content/github_aistream-peelout_flow-forecast')
!python setup.py develop
!pip install -r requirements.txt
!mkdir data
from flood_forecast.trainer import train_function
!pip install git+https://github.com/CoronaWhy/task-geo.git
!pip install git+https://github.com/coronawhy/task-ts
import wandb
/bin/bash: wandb: command not found WARNING: Repository "github_aistream-peelout_flow-forecast" in project "gmap-997" is a mirror. Pushing to this clone will have no effect. Instead, clone the mirrored repository directly with $ git clone https://github.com/AIStream-Peelout/flow-forecast Cloning into '/content/github_aistream-peelout_flow-forecast'... remote: Total 3943 (delta 2553), reused 3943 (delta 2553) Receiving objects: 100% (3943/3943), 2.71 MiB | 15.67 MiB/s, done. Resolving deltas: 100% (2553/2553), done. Project [gmap-997] repository [github_aistream-peelout_flow-forecast] was cloned to [/content/github_aistream-peelout_flow-forecast]. /usr/local/lib/python3.6/dist-packages/setuptools/dist.py:454: UserWarning: Normalizing '0.01dev' to '0.1.dev0' warnings.warn(tmpl.format(**locals())) running develop running egg_info creating flood_forecast.egg-info writing flood_forecast.egg-info/PKG-INFO writing dependency_links to flood_forecast.egg-info/dependency_links.txt writing requirements to flood_forecast.egg-info/requires.txt writing top-level names to flood_forecast.egg-info/top_level.txt writing manifest file 'flood_forecast.egg-info/SOURCES.txt' package init file 'flood_forecast/__init__.py' not found (or not a regular file) package init file 'flood_forecast/transformer_xl/__init__.py' not found (or not a regular file) package init file 'flood_forecast/preprocessing/__init__.py' not found (or not a regular file) package init file 'flood_forecast/da_rnn/__init__.py' not found (or not a regular file) package init file 'flood_forecast/basic/__init__.py' not found (or not a regular file) package init file 'flood_forecast/custom/__init__.py' not found (or not a regular file) writing manifest file 'flood_forecast.egg-info/SOURCES.txt' running build_ext Creating /usr/local/lib/python3.6/dist-packages/flood-forecast.egg-link (link to .) Adding flood-forecast 0.1.dev0 to easy-install.pth file Installed /content/github_aistream-peelout_flow-forecast Processing dependencies for flood-forecast==0.1.dev0 Searching for google-cloud Reading https://pypi.org/simple/google-cloud/ Downloading https://files.pythonhosted.org/packages/ba/b1/7c54d1950e7808df06642274e677dbcedba57f75307adf2e5ad8d39e5e0e/google_cloud-0.34.0-py2.py3-none-any.whl#sha256=fb1ab7b0548fe44b3d538041f0a374505b7f990d448a935ea36649c5ccab5acf Best match: google-cloud 0.34.0 Processing google_cloud-0.34.0-py2.py3-none-any.whl Installing google_cloud-0.34.0-py2.py3-none-any.whl to /usr/local/lib/python3.6/dist-packages Adding google-cloud 0.34.0 to easy-install.pth file Installed /usr/local/lib/python3.6/dist-packages/google_cloud-0.34.0-py3.6.egg Searching for pandas==1.0.3 Best match: pandas 1.0.3 Adding pandas 1.0.3 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for tensorflow==2.2.0 Best match: tensorflow 2.2.0 Adding tensorflow 2.2.0 to easy-install.pth file Installing estimator_ckpt_converter script to /usr/local/bin Installing saved_model_cli script to /usr/local/bin Installing tensorboard script to /usr/local/bin Installing tf_upgrade_v2 script to /usr/local/bin Installing tflite_convert script to /usr/local/bin Installing toco script to /usr/local/bin Installing toco_from_protos script to /usr/local/bin Using /usr/local/lib/python3.6/dist-packages Searching for torch==1.5.0+cu101 Best match: torch 1.5.0+cu101 Adding torch 1.5.0+cu101 to easy-install.pth file Installing convert-caffe2-to-onnx script to /usr/local/bin Installing convert-onnx-to-caffe2 script to /usr/local/bin Using /usr/local/lib/python3.6/dist-packages Searching for scikit-learn==0.22.2.post1 Best match: scikit-learn 0.22.2.post1 Adding scikit-learn 0.22.2.post1 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for numpy==1.18.4 Best match: numpy 1.18.4 Adding numpy 1.18.4 to easy-install.pth file Installing f2py script to /usr/local/bin Installing f2py3 script to /usr/local/bin Installing f2py3.6 script to /usr/local/bin Using /usr/local/lib/python3.6/dist-packages Searching for python-dateutil==2.8.1 Best match: python-dateutil 2.8.1 Adding python-dateutil 2.8.1 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for pytz==2018.9 Best match: pytz 2018.9 Adding pytz 2018.9 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for termcolor==1.1.0 Best match: termcolor 1.1.0 Adding termcolor 1.1.0 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for opt-einsum==3.2.1 Best match: opt-einsum 3.2.1 Adding opt-einsum 3.2.1 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for absl-py==0.9.0 Best match: absl-py 0.9.0 Adding absl-py 0.9.0 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for wheel==0.34.2 Best match: wheel 0.34.2 Adding wheel 0.34.2 to easy-install.pth file Installing wheel script to /usr/local/bin Using /usr/local/lib/python3.6/dist-packages Searching for protobuf==3.10.0 Best match: protobuf 3.10.0 Adding protobuf 3.10.0 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for tensorflow-estimator==2.2.0 Best match: tensorflow-estimator 2.2.0 Adding tensorflow-estimator 2.2.0 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for h5py==2.10.0 Best match: h5py 2.10.0 Adding h5py 2.10.0 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for scipy==1.4.1 Best match: scipy 1.4.1 Adding scipy 1.4.1 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for wrapt==1.12.1 Best match: wrapt 1.12.1 Adding wrapt 1.12.1 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for astunparse==1.6.3 Best match: astunparse 1.6.3 Adding astunparse 1.6.3 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for six==1.12.0 Best match: six 1.12.0 Adding six 1.12.0 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for Keras-Preprocessing==1.1.2 Best match: Keras-Preprocessing 1.1.2 Adding Keras-Preprocessing 1.1.2 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for tensorboard==2.2.1 Best match: tensorboard 2.2.1 Adding tensorboard 2.2.1 to easy-install.pth file Installing tensorboard script to /usr/local/bin Using /usr/local/lib/python3.6/dist-packages Searching for grpcio==1.29.0 Best match: grpcio 1.29.0 Adding grpcio 1.29.0 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for gast==0.3.3 Best match: gast 0.3.3 Adding gast 0.3.3 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for google-pasta==0.2.0 Best match: google-pasta 0.2.0 Adding google-pasta 0.2.0 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for future==0.16.0 Best match: future 0.16.0 Adding future 0.16.0 to easy-install.pth file Installing futurize script to /usr/local/bin Installing pasteurize script to /usr/local/bin Using /usr/local/lib/python3.6/dist-packages Searching for joblib==0.15.1 Best match: joblib 0.15.1 Adding joblib 0.15.1 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for setuptools==46.4.0 Best match: setuptools 46.4.0 Adding setuptools 46.4.0 to easy-install.pth file Installing easy_install script to /usr/local/bin Installing easy_install-3.8 script to /usr/local/bin Using /usr/local/lib/python3.6/dist-packages Searching for Markdown==3.2.2 Best match: Markdown 3.2.2 Adding Markdown 3.2.2 to easy-install.pth file Installing markdown_py script to /usr/local/bin Using /usr/local/lib/python3.6/dist-packages Searching for tensorboard-plugin-wit==1.6.0.post3 Best match: tensorboard-plugin-wit 1.6.0.post3 Adding tensorboard-plugin-wit 1.6.0.post3 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for google-auth==1.7.2 Best match: google-auth 1.7.2 Adding google-auth 1.7.2 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for google-auth-oauthlib==0.4.1 Best match: google-auth-oauthlib 0.4.1 Adding google-auth-oauthlib 0.4.1 to easy-install.pth file Installing google-oauthlib-tool script to /usr/local/bin Using /usr/local/lib/python3.6/dist-packages Searching for requests==2.23.0 Best match: requests 2.23.0 Adding requests 2.23.0 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for Werkzeug==1.0.1 Best match: Werkzeug 1.0.1 Adding Werkzeug 1.0.1 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for importlib-metadata==1.6.0 Best match: importlib-metadata 1.6.0 Adding importlib-metadata 1.6.0 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for pyasn1-modules==0.2.8 Best match: pyasn1-modules 0.2.8 Adding pyasn1-modules 0.2.8 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for rsa==4.0 Best match: rsa 4.0 Adding rsa 4.0 to easy-install.pth file Installing pyrsa-decrypt script to /usr/local/bin Installing pyrsa-encrypt script to /usr/local/bin Installing pyrsa-keygen script to /usr/local/bin Installing pyrsa-priv2pub script to /usr/local/bin Installing pyrsa-sign script to /usr/local/bin Installing pyrsa-verify script to /usr/local/bin Using /usr/local/lib/python3.6/dist-packages Searching for cachetools==3.1.1 Best match: cachetools 3.1.1 Adding cachetools 3.1.1 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for requests-oauthlib==1.3.0 Best match: requests-oauthlib 1.3.0 Adding requests-oauthlib 1.3.0 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for urllib3==1.24.3 Best match: urllib3 1.24.3 Adding urllib3 1.24.3 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for chardet==3.0.4 Best match: chardet 3.0.4 Adding chardet 3.0.4 to easy-install.pth file Installing chardetect script to /usr/local/bin Using /usr/local/lib/python3.6/dist-packages Searching for idna==2.9 Best match: idna 2.9 Adding idna 2.9 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for certifi==2020.4.5.1 Best match: certifi 2020.4.5.1 Adding certifi 2020.4.5.1 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for zipp==3.1.0 Best match: zipp 3.1.0 Adding zipp 3.1.0 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for pyasn1==0.4.8 Best match: pyasn1 0.4.8 Adding pyasn1 0.4.8 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Searching for oauthlib==3.1.0 Best match: oauthlib 3.1.0 Adding oauthlib 3.1.0 to easy-install.pth file Using /usr/local/lib/python3.6/dist-packages Finished processing dependencies for flood-forecast==0.1.dev0 Requirement already satisfied: scikit-learn in /usr/local/lib/python3.6/dist-packages (from -r requirements.txt (line 1)) (0.22.2.post1) Requirement already satisfied: pandas in /usr/local/lib/python3.6/dist-packages (from -r requirements.txt (line 2)) (1.0.3) Requirement already satisfied: torch in /usr/local/lib/python3.6/dist-packages (from -r requirements.txt (line 3)) (1.5.0+cu101) Collecting tb-nightly Downloading https://files.pythonhosted.org/packages/cd/85/17daa51272812331873571020632fc0f5300f0d448e51ec584a7bacc0fa7/tb_nightly-2.3.0a20200528-py3-none-any.whl (2.9MB) |████████████████████████████████| 2.9MB 1.7MB/s Requirement already satisfied: seaborn in /usr/local/lib/python3.6/dist-packages (from -r requirements.txt (line 5)) (0.10.1) Requirement already satisfied: future in /usr/local/lib/python3.6/dist-packages (from -r requirements.txt (line 6)) (0.16.0) Collecting wandb Downloading https://files.pythonhosted.org/packages/d1/c7/8bf2c62c3f133f45e135a8a116e4e0f162043248e3db54de30996eaf1a8a/wandb-0.8.36-py2.py3-none-any.whl (1.4MB) |████████████████████████████████| 1.4MB 18.1MB/s Requirement already satisfied: pandas_gbq in /usr/local/lib/python3.6/dist-packages (from -r requirements.txt (line 8)) (0.11.0) Requirement already satisfied: google-cloud in /usr/local/lib/python3.6/dist-packages/google_cloud-0.34.0-py3.6.egg (from -r requirements.txt (line 9)) (0.34.0) Requirement already satisfied: google-cloud-storage in /usr/local/lib/python3.6/dist-packages (from -r requirements.txt (line 10)) (1.18.1) Requirement already satisfied: pyyaml in /usr/local/lib/python3.6/dist-packages (from -r requirements.txt (line 11)) (3.13) Requirement already satisfied: plotly in /usr/local/lib/python3.6/dist-packages (from -r requirements.txt (line 12)) (4.4.1) Requirement already satisfied: scipy>=0.17.0 in /usr/local/lib/python3.6/dist-packages (from scikit-learn->-r requirements.txt (line 1)) (1.4.1) Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.6/dist-packages (from scikit-learn->-r requirements.txt (line 1)) (0.15.1) Requirement already satisfied: numpy>=1.11.0 in /usr/local/lib/python3.6/dist-packages (from scikit-learn->-r requirements.txt (line 1)) (1.18.4) Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.6/dist-packages (from pandas->-r requirements.txt (line 2)) (2018.9) Requirement already satisfied: python-dateutil>=2.6.1 in /usr/local/lib/python3.6/dist-packages (from pandas->-r requirements.txt (line 2)) (2.8.1) Requirement already satisfied: protobuf>=3.6.0 in /usr/local/lib/python3.6/dist-packages (from tb-nightly->-r requirements.txt (line 4)) (3.10.0) Requirement already satisfied: tensorboard-plugin-wit>=1.6.0 in /usr/local/lib/python3.6/dist-packages (from tb-nightly->-r requirements.txt (line 4)) (1.6.0.post3) Requirement already satisfied: absl-py>=0.4 in /usr/local/lib/python3.6/dist-packages (from tb-nightly->-r requirements.txt (line 4)) (0.9.0) Requirement already satisfied: requests<3,>=2.21.0 in /usr/local/lib/python3.6/dist-packages (from tb-nightly->-r requirements.txt (line 4)) (2.23.0) Requirement already satisfied: six>=1.10.0 in /usr/local/lib/python3.6/dist-packages (from tb-nightly->-r requirements.txt (line 4)) (1.12.0) Requirement already satisfied: werkzeug>=0.11.15 in /usr/local/lib/python3.6/dist-packages (from tb-nightly->-r requirements.txt (line 4)) (1.0.1) Requirement already satisfied: wheel>=0.26; python_version >= "3" in /usr/local/lib/python3.6/dist-packages (from tb-nightly->-r requirements.txt (line 4)) (0.34.2) Requirement already satisfied: google-auth<2,>=1.6.3 in /usr/local/lib/python3.6/dist-packages (from tb-nightly->-r requirements.txt (line 4)) (1.7.2) Requirement already satisfied: google-auth-oauthlib<0.5,>=0.4.1 in /usr/local/lib/python3.6/dist-packages (from tb-nightly->-r requirements.txt (line 4)) (0.4.1) Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.6/dist-packages (from tb-nightly->-r requirements.txt (line 4)) (3.2.2) Requirement already satisfied: grpcio>=1.24.3 in /usr/local/lib/python3.6/dist-packages (from tb-nightly->-r requirements.txt (line 4)) (1.29.0) Requirement already satisfied: setuptools>=41.0.0 in /usr/local/lib/python3.6/dist-packages (from tb-nightly->-r requirements.txt (line 4)) (46.4.0) Requirement already satisfied: matplotlib>=2.1.2 in /usr/local/lib/python3.6/dist-packages (from seaborn->-r requirements.txt (line 5)) (3.2.1) Collecting GitPython>=1.0.0 Downloading https://files.pythonhosted.org/packages/44/33/917e6fde1cad13daa7053f39b7c8af3be287314f75f1b1ea8d3fe37a8571/GitPython-3.1.2-py3-none-any.whl (451kB) |████████████████████████████████| 460kB 15.3MB/s Collecting subprocess32>=3.5.3 Downloading https://files.pythonhosted.org/packages/32/c8/564be4d12629b912ea431f1a50eb8b3b9d00f1a0b1ceff17f266be190007/subprocess32-3.5.4.tar.gz (97kB) |████████████████████████████████| 102kB 9.5MB/s Collecting watchdog>=0.8.3 Downloading https://files.pythonhosted.org/packages/73/c3/ed6d992006837e011baca89476a4bbffb0a91602432f73bd4473816c76e2/watchdog-0.10.2.tar.gz (95kB) |████████████████████████████████| 102kB 9.2MB/s Collecting docker-pycreds>=0.4.0 Downloading https://files.pythonhosted.org/packages/f5/e8/f6bd1eee09314e7e6dee49cbe2c5e22314ccdb38db16c9fc72d2fa80d054/docker_pycreds-0.4.0-py2.py3-none-any.whl Collecting configparser>=3.8.1 Downloading https://files.pythonhosted.org/packages/4b/6b/01baa293090240cf0562cc5eccb69c6f5006282127f2b846fad011305c79/configparser-5.0.0-py3-none-any.whl Requirement already satisfied: Click>=7.0 in /usr/local/lib/python3.6/dist-packages (from wandb->-r requirements.txt (line 7)) (7.1.2) Collecting shortuuid>=0.5.0 Downloading https://files.pythonhosted.org/packages/25/a6/2ecc1daa6a304e7f1b216f0896b26156b78e7c38e1211e9b798b4716c53d/shortuuid-1.0.1-py3-none-any.whl Requirement already satisfied: nvidia-ml-py3>=7.352.0 in /usr/local/lib/python3.6/dist-packages (from wandb->-r requirements.txt (line 7)) (7.352.0) Collecting sentry-sdk>=0.4.0 Downloading https://files.pythonhosted.org/packages/1b/95/9a20eebcedab2c1c63fad59fe19a0469edfc2a25b8576497e8084629c2ff/sentry_sdk-0.14.4-py2.py3-none-any.whl (104kB) |████████████████████████████████| 112kB 24.1MB/s Collecting gql==0.2.0 Downloading https://files.pythonhosted.org/packages/c4/6f/cf9a3056045518f06184e804bae89390eb706168349daa9dff8ac609962a/gql-0.2.0.tar.gz Requirement already satisfied: psutil>=5.0.0 in /usr/local/lib/python3.6/dist-packages (from wandb->-r requirements.txt (line 7)) (5.4.8) Requirement already satisfied: pydata-google-auth in /usr/local/lib/python3.6/dist-packages (from pandas_gbq->-r requirements.txt (line 8)) (1.1.0) Requirement already satisfied: google-cloud-bigquery>=1.9.0 in /usr/local/lib/python3.6/dist-packages (from pandas_gbq->-r requirements.txt (line 8)) (1.21.0) Requirement already satisfied: google-resumable-media<0.5.0dev,>=0.3.1 in /usr/local/lib/python3.6/dist-packages (from google-cloud-storage->-r requirements.txt (line 10)) (0.4.1) Requirement already satisfied: google-cloud-core<2.0dev,>=1.0.0 in /usr/local/lib/python3.6/dist-packages (from google-cloud-storage->-r requirements.txt (line 10)) (1.0.3) Requirement already satisfied: retrying>=1.3.3 in /usr/local/lib/python3.6/dist-packages (from plotly->-r requirements.txt (line 12)) (1.3.3) Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests<3,>=2.21.0->tb-nightly->-r requirements.txt (line 4)) (3.0.4) Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests<3,>=2.21.0->tb-nightly->-r requirements.txt (line 4)) (2.9) Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests<3,>=2.21.0->tb-nightly->-r requirements.txt (line 4)) (2020.4.5.1) Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests<3,>=2.21.0->tb-nightly->-r requirements.txt (line 4)) (1.24.3) Requirement already satisfied: cachetools<3.2,>=2.0.0 in /usr/local/lib/python3.6/dist-packages (from google-auth<2,>=1.6.3->tb-nightly->-r requirements.txt (line 4)) (3.1.1) Requirement already satisfied: rsa<4.1,>=3.1.4 in /usr/local/lib/python3.6/dist-packages (from google-auth<2,>=1.6.3->tb-nightly->-r requirements.txt (line 4)) (4.0) Requirement already satisfied: pyasn1-modules>=0.2.1 in /usr/local/lib/python3.6/dist-packages (from google-auth<2,>=1.6.3->tb-nightly->-r requirements.txt (line 4)) (0.2.8) Requirement already satisfied: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.6/dist-packages (from google-auth-oauthlib<0.5,>=0.4.1->tb-nightly->-r requirements.txt (line 4)) (1.3.0) Requirement already satisfied: importlib-metadata; python_version < "3.8" in /usr/local/lib/python3.6/dist-packages (from markdown>=2.6.8->tb-nightly->-r requirements.txt (line 4)) (1.6.0) Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=2.1.2->seaborn->-r requirements.txt (line 5)) (0.10.0) Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=2.1.2->seaborn->-r requirements.txt (line 5)) (2.4.7) Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=2.1.2->seaborn->-r requirements.txt (line 5)) (1.2.0) Collecting gitdb<5,>=4.0.1 Downloading https://files.pythonhosted.org/packages/48/11/d1800bca0a3bae820b84b7d813ad1eff15a48a64caea9c823fc8c1b119e8/gitdb-4.0.5-py3-none-any.whl (63kB) |████████████████████████████████| 71kB 8.9MB/s Collecting pathtools>=0.1.1 Downloading https://files.pythonhosted.org/packages/e7/7f/470d6fcdf23f9f3518f6b0b76be9df16dcc8630ad409947f8be2eb0ed13a/pathtools-0.1.2.tar.gz Collecting graphql-core<2,>=0.5.0 Downloading https://files.pythonhosted.org/packages/b0/89/00ad5e07524d8c523b14d70c685e0299a8b0de6d0727e368c41b89b7ed0b/graphql-core-1.1.tar.gz (70kB) |████████████████████████████████| 71kB 8.2MB/s Requirement already satisfied: promise<3,>=2.0 in /usr/local/lib/python3.6/dist-packages (from gql==0.2.0->wandb->-r requirements.txt (line 7)) (2.3) Requirement already satisfied: google-api-core<2.0.0dev,>=1.14.0 in /usr/local/lib/python3.6/dist-packages (from google-cloud-core<2.0dev,>=1.0.0->google-cloud-storage->-r requirements.txt (line 10)) (1.16.0) Requirement already satisfied: pyasn1>=0.1.3 in /usr/local/lib/python3.6/dist-packages (from rsa<4.1,>=3.1.4->google-auth<2,>=1.6.3->tb-nightly->-r requirements.txt (line 4)) (0.4.8) Requirement already satisfied: oauthlib>=3.0.0 in /usr/local/lib/python3.6/dist-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<0.5,>=0.4.1->tb-nightly->-r requirements.txt (line 4)) (3.1.0) Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.6/dist-packages (from importlib-metadata; python_version < "3.8"->markdown>=2.6.8->tb-nightly->-r requirements.txt (line 4)) (3.1.0) Collecting smmap<4,>=3.0.1 Downloading https://files.pythonhosted.org/packages/b0/9a/4d409a6234eb940e6a78dfdfc66156e7522262f5f2fecca07dc55915952d/smmap-3.0.4-py2.py3-none-any.whl Requirement already satisfied: googleapis-common-protos<2.0dev,>=1.6.0 in /usr/local/lib/python3.6/dist-packages (from google-api-core<2.0.0dev,>=1.14.0->google-cloud-core<2.0dev,>=1.0.0->google-cloud-storage->-r requirements.txt (line 10)) (1.51.0) Building wheels for collected packages: subprocess32, watchdog, gql, pathtools, graphql-core Building wheel for subprocess32 (setup.py) ... done Created wheel for subprocess32: filename=subprocess32-3.5.4-cp36-none-any.whl size=6489 sha256=280884442d07c60651c8f68f55c26b34ba44f7d807411e2be43187490e7c3c40 Stored in directory: /root/.cache/pip/wheels/68/39/1a/5e402bdfdf004af1786c8b853fd92f8c4a04f22aad179654d1 Building wheel for watchdog (setup.py) ... done Created wheel for watchdog: filename=watchdog-0.10.2-cp36-none-any.whl size=73605 sha256=48a49c4a1d738f6eb0135ce1fd14986d344660098beba8ef979530d4cfbacde8 Stored in directory: /root/.cache/pip/wheels/bc/ed/6c/028dea90d31b359cd2a7c8b0da4db80e41d24a59614154072e Building wheel for gql (setup.py) ... done Created wheel for gql: filename=gql-0.2.0-cp36-none-any.whl size=7630 sha256=38548a3896ff23d13edf364c5125fa4da8dfef43586b0bc0971fc8077a8cbb10 Stored in directory: /root/.cache/pip/wheels/ce/0e/7b/58a8a5268655b3ad74feef5aa97946f0addafb3cbb6bd2da23 Building wheel for pathtools (setup.py) ... done Created wheel for pathtools: filename=pathtools-0.1.2-cp36-none-any.whl size=8784 sha256=3a2eb30a5b42227e673fc0f8b1098f1536659c38f7d04bbeb1a12fa11005067d Stored in directory: /root/.cache/pip/wheels/0b/04/79/c3b0c3a0266a3cb4376da31e5bfe8bba0c489246968a68e843 Building wheel for graphql-core (setup.py) ... done Created wheel for graphql-core: filename=graphql_core-1.1-cp36-none-any.whl size=104650 sha256=a48d17e184fc63251279a4b06e80d7f46a23cd63a0e447b862166f0a1ae83a7e Stored in directory: /root/.cache/pip/wheels/45/99/d7/c424029bb0fe910c63b68dbf2aa20d3283d023042521bcd7d5 Successfully built subprocess32 watchdog gql pathtools graphql-core Installing collected packages: tb-nightly, smmap, gitdb, GitPython, subprocess32, pathtools, watchdog, docker-pycreds, configparser, shortuuid, sentry-sdk, graphql-core, gql, wandb Successfully installed GitPython-3.1.2 configparser-5.0.0 docker-pycreds-0.4.0 gitdb-4.0.5 gql-0.2.0 graphql-core-1.1 pathtools-0.1.2 sentry-sdk-0.14.4 shortuuid-1.0.1 smmap-3.0.4 subprocess32-3.5.4 tb-nightly-2.3.0a20200528 wandb-0.8.36 watchdog-0.10.2 Collecting git+https://github.com/CoronaWhy/task-geo.git Cloning https://github.com/CoronaWhy/task-geo.git to /tmp/pip-req-build-x2yg5c0b Running command git clone -q https://github.com/CoronaWhy/task-geo.git /tmp/pip-req-build-x2yg5c0b Requirement already satisfied: pandas in /usr/local/lib/python3.6/dist-packages (from task-geo==0.1.0.dev0) (1.0.3) Requirement already satisfied: requests in /usr/local/lib/python3.6/dist-packages (from task-geo==0.1.0.dev0) (2.23.0) Requirement already satisfied: jupyter in /usr/local/lib/python3.6/dist-packages (from task-geo==0.1.0.dev0) (1.0.0) Collecting hdx-python-api Downloading https://files.pythonhosted.org/packages/79/32/7033d6d9ff01fd592ec649756f78460dc66640c1001f39c2d421037866f3/hdx_python_api-4.5.8-py2.py3-none-any.whl (67kB) |████████████████████████████████| 71kB 1.6MB/s Requirement already satisfied: numpy>=1.13.3 in /usr/local/lib/python3.6/dist-packages (from pandas->task-geo==0.1.0.dev0) (1.18.4) Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.6/dist-packages (from pandas->task-geo==0.1.0.dev0) (2018.9) Requirement already satisfied: python-dateutil>=2.6.1 in /usr/local/lib/python3.6/dist-packages (from pandas->task-geo==0.1.0.dev0) (2.8.1) Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests->task-geo==0.1.0.dev0) (2020.4.5.1) Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests->task-geo==0.1.0.dev0) (2.9) Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests->task-geo==0.1.0.dev0) (1.24.3) Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests->task-geo==0.1.0.dev0) (3.0.4) Requirement already satisfied: nbconvert in /usr/local/lib/python3.6/dist-packages (from jupyter->task-geo==0.1.0.dev0) (5.6.1) Requirement already satisfied: ipykernel in /usr/local/lib/python3.6/dist-packages (from jupyter->task-geo==0.1.0.dev0) (4.10.1) Requirement already satisfied: qtconsole in /usr/local/lib/python3.6/dist-packages (from jupyter->task-geo==0.1.0.dev0) (4.7.4) Requirement already satisfied: jupyter-console in /usr/local/lib/python3.6/dist-packages (from jupyter->task-geo==0.1.0.dev0) (5.2.0) Requirement already satisfied: notebook in /usr/local/lib/python3.6/dist-packages (from jupyter->task-geo==0.1.0.dev0) (5.2.2) Requirement already satisfied: ipywidgets in /usr/local/lib/python3.6/dist-packages (from jupyter->task-geo==0.1.0.dev0) (7.5.1) Collecting ckanapi>=4.3 Downloading https://files.pythonhosted.org/packages/07/c6/50449e18aaf1600dfda955805c58aa7462493511f3ebbb20d0a65874397c/ckanapi-4.3.tar.gz Collecting pyOpenSSL Downloading https://files.pythonhosted.org/packages/9e/de/f8342b68fa9e981d348039954657bdf681b2ab93de27443be51865ffa310/pyOpenSSL-19.1.0-py2.py3-none-any.whl (53kB) |████████████████████████████████| 61kB 5.6MB/s Requirement already satisfied: pyasn1 in /usr/local/lib/python3.6/dist-packages (from hdx-python-api->task-geo==0.1.0.dev0) (0.4.8) Collecting hdx-python-country>=2.5.6 Downloading https://files.pythonhosted.org/packages/24/45/684ec41e237b3e185c268b4ad4cd71581c3dfdb9c3e279d7777473edb8c2/hdx_python_country-2.5.6-py2.py3-none-any.whl Collecting ndg-httpsclient Downloading https://files.pythonhosted.org/packages/fb/67/c2f508c00ed2a6911541494504b7cac16fe0b0473912568df65fd1801132/ndg_httpsclient-0.5.1-py3-none-any.whl Collecting quantulum3>=0.7.3; python_version >= "3" Downloading https://files.pythonhosted.org/packages/d0/d5/60d72e7c9393199eba72f88deb06e0fddc4e357d600bdbe5f86e8cfc4512/quantulum3-0.7.3-py3-none-any.whl (11.1MB) |████████████████████████████████| 11.1MB 5.3MB/s Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.6/dist-packages (from python-dateutil>=2.6.1->pandas->task-geo==0.1.0.dev0) (1.12.0) Requirement already satisfied: pygments in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo==0.1.0.dev0) (2.1.3) Requirement already satisfied: nbformat>=4.4 in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo==0.1.0.dev0) (5.0.6) Requirement already satisfied: bleach in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo==0.1.0.dev0) (3.1.5) Requirement already satisfied: traitlets>=4.2 in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo==0.1.0.dev0) (4.3.3) Requirement already satisfied: mistune<2,>=0.8.1 in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo==0.1.0.dev0) (0.8.4) Requirement already satisfied: pandocfilters>=1.4.1 in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo==0.1.0.dev0) (1.4.2) Requirement already satisfied: defusedxml in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo==0.1.0.dev0) (0.6.0) Requirement already satisfied: jupyter-core in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo==0.1.0.dev0) (4.6.3) Requirement already satisfied: jinja2>=2.4 in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo==0.1.0.dev0) (2.11.2) Requirement already satisfied: testpath in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo==0.1.0.dev0) (0.4.4) Requirement already satisfied: entrypoints>=0.2.2 in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo==0.1.0.dev0) (0.3) Requirement already satisfied: ipython>=4.0.0 in /usr/local/lib/python3.6/dist-packages (from ipykernel->jupyter->task-geo==0.1.0.dev0) (5.5.0) Requirement already satisfied: jupyter-client in /usr/local/lib/python3.6/dist-packages (from ipykernel->jupyter->task-geo==0.1.0.dev0) (5.3.4) Requirement already satisfied: tornado>=4.0 in /usr/local/lib/python3.6/dist-packages (from ipykernel->jupyter->task-geo==0.1.0.dev0) (4.5.3) Requirement already satisfied: qtpy in /usr/local/lib/python3.6/dist-packages (from qtconsole->jupyter->task-geo==0.1.0.dev0) (1.9.0) Requirement already satisfied: pyzmq>=17.1 in /usr/local/lib/python3.6/dist-packages (from qtconsole->jupyter->task-geo==0.1.0.dev0) (19.0.1) Requirement already satisfied: ipython-genutils in /usr/local/lib/python3.6/dist-packages (from qtconsole->jupyter->task-geo==0.1.0.dev0) (0.2.0) Requirement already satisfied: prompt-toolkit<2.0.0,>=1.0.0 in /usr/local/lib/python3.6/dist-packages (from jupyter-console->jupyter->task-geo==0.1.0.dev0) (1.0.18) Requirement already satisfied: terminado>=0.3.3; sys_platform != "win32" in /usr/local/lib/python3.6/dist-packages (from notebook->jupyter->task-geo==0.1.0.dev0) (0.8.3) Requirement already satisfied: widgetsnbextension~=3.5.0 in /usr/local/lib/python3.6/dist-packages (from ipywidgets->jupyter->task-geo==0.1.0.dev0) (3.5.1) Requirement already satisfied: setuptools in /usr/local/lib/python3.6/dist-packages (from ckanapi>=4.3->hdx-python-api->task-geo==0.1.0.dev0) (46.4.0) Requirement already satisfied: docopt in /usr/local/lib/python3.6/dist-packages (from ckanapi>=4.3->hdx-python-api->task-geo==0.1.0.dev0) (0.6.2) Requirement already satisfied: python-slugify>=1.0 in /usr/local/lib/python3.6/dist-packages (from ckanapi>=4.3->hdx-python-api->task-geo==0.1.0.dev0) (4.0.0) Collecting cryptography>=2.8 Downloading https://files.pythonhosted.org/packages/3c/04/686efee2dcdd25aecf357992e7d9362f443eb182ecd623f882bc9f7a6bba/cryptography-2.9.2-cp35-abi3-manylinux2010_x86_64.whl (2.7MB) |████████████████████████████████| 2.7MB 42.0MB/s Collecting libhxl>=4.19; python_version >= "3" Downloading https://files.pythonhosted.org/packages/3d/4e/e47c858f5a3bc3b21d76d4dd8a85cc1cf876ca8ab68fe734f37cb504a685/libhxl-4.20.tar.gz (83kB) |████████████████████████████████| 92kB 11.6MB/s Collecting hdx-python-utilities>=2.3.4 Downloading https://files.pythonhosted.org/packages/13/0c/dd417c960de745a129d6c0cf63bb0215b179c29674e45ca041c75cf76baf/hdx_python_utilities-2.3.4-py2.py3-none-any.whl (46kB) |████████████████████████████████| 51kB 7.2MB/s Collecting num2words Downloading https://files.pythonhosted.org/packages/eb/a2/ea800689730732e27711c41beed4b2a129b34974435bdc450377ec407738/num2words-0.5.10-py3-none-any.whl (101kB) |████████████████████████████████| 102kB 11.9MB/s Requirement already satisfied: inflect in /usr/local/lib/python3.6/dist-packages (from quantulum3>=0.7.3; python_version >= "3"->hdx-python-api->task-geo==0.1.0.dev0) (2.1.0) Requirement already satisfied: jsonschema!=2.5.0,>=2.4 in /usr/local/lib/python3.6/dist-packages (from nbformat>=4.4->nbconvert->jupyter->task-geo==0.1.0.dev0) (2.6.0) Requirement already satisfied: packaging in /usr/local/lib/python3.6/dist-packages (from bleach->nbconvert->jupyter->task-geo==0.1.0.dev0) (20.4) Requirement already satisfied: webencodings in /usr/local/lib/python3.6/dist-packages (from bleach->nbconvert->jupyter->task-geo==0.1.0.dev0) (0.5.1) Requirement already satisfied: decorator in /usr/local/lib/python3.6/dist-packages (from traitlets>=4.2->nbconvert->jupyter->task-geo==0.1.0.dev0) (4.4.2) Requirement already satisfied: MarkupSafe>=0.23 in /usr/local/lib/python3.6/dist-packages (from jinja2>=2.4->nbconvert->jupyter->task-geo==0.1.0.dev0) (1.1.1) Requirement already satisfied: pickleshare in /usr/local/lib/python3.6/dist-packages (from ipython>=4.0.0->ipykernel->jupyter->task-geo==0.1.0.dev0) (0.7.5) Requirement already satisfied: simplegeneric>0.8 in /usr/local/lib/python3.6/dist-packages (from ipython>=4.0.0->ipykernel->jupyter->task-geo==0.1.0.dev0) (0.8.1) Requirement already satisfied: pexpect; sys_platform != "win32" in /usr/local/lib/python3.6/dist-packages (from ipython>=4.0.0->ipykernel->jupyter->task-geo==0.1.0.dev0) (4.8.0) Requirement already satisfied: wcwidth in /usr/local/lib/python3.6/dist-packages (from prompt-toolkit<2.0.0,>=1.0.0->jupyter-console->jupyter->task-geo==0.1.0.dev0) (0.1.9) Requirement already satisfied: ptyprocess; os_name != "nt" in /usr/local/lib/python3.6/dist-packages (from terminado>=0.3.3; sys_platform != "win32"->notebook->jupyter->task-geo==0.1.0.dev0) (0.6.0) Requirement already satisfied: text-unidecode>=1.3 in /usr/local/lib/python3.6/dist-packages (from python-slugify>=1.0->ckanapi>=4.3->hdx-python-api->task-geo==0.1.0.dev0) (1.3) Requirement already satisfied: cffi!=1.11.3,>=1.8 in /usr/local/lib/python3.6/dist-packages (from cryptography>=2.8->pyOpenSSL->hdx-python-api->task-geo==0.1.0.dev0) (1.14.0) Requirement already satisfied: xlrd in /usr/local/lib/python3.6/dist-packages (from libhxl>=4.19; python_version >= "3"->hdx-python-country>=2.5.6->hdx-python-api->task-geo==0.1.0.dev0) (1.1.0) Collecting unidecode Downloading https://files.pythonhosted.org/packages/d0/42/d9edfed04228bacea2d824904cae367ee9efd05e6cce7ceaaedd0b0ad964/Unidecode-1.1.1-py2.py3-none-any.whl (238kB) |████████████████████████████████| 245kB 41.2MB/s Collecting python-io-wrapper Downloading https://files.pythonhosted.org/packages/76/81/88e02bc603e55883a087811a641fd3836749b7509365778fea29d74fd58c/python-io-wrapper-0.1.tar.gz Collecting jsonpath_ng Downloading https://files.pythonhosted.org/packages/85/24/981fa76e1e415bcceb3a9741d5be04b32c10edd42d88f6783faa750b1239/jsonpath-ng-1.5.1.tar.gz Collecting ply Downloading https://files.pythonhosted.org/packages/a3/58/35da89ee790598a0700ea49b2a66594140f44dec458c07e8e3d4979137fc/ply-3.11-py2.py3-none-any.whl (49kB) |████████████████████████████████| 51kB 7.2MB/s Collecting ratelimit Downloading https://files.pythonhosted.org/packages/ab/38/ff60c8fc9e002d50d48822cc5095deb8ebbc5f91a6b8fdd9731c87a147c9/ratelimit-2.2.1.tar.gz Collecting psycopg2-binary Downloading https://files.pythonhosted.org/packages/d3/8a/a7ed55c2c55bd4f5844d72734fedc0cef8a74518a0a19105a21c15628f1e/psycopg2_binary-2.8.5-cp36-cp36m-manylinux1_x86_64.whl (2.9MB) |████████████████████████████████| 2.9MB 46.3MB/s Collecting yamlloader Downloading https://files.pythonhosted.org/packages/93/a2/2f0c2394af1559021703c8cbb1bc7419bb5a94ea6bde0ab8cd1e973bb605/yamlloader-0.5.5-py3-none-any.whl Collecting sshtunnel Downloading https://files.pythonhosted.org/packages/c5/5c/4b320d7ec4b0d5d4d6df1fdf66a5799625b3623d0ce4efe81719c6f8dfb3/sshtunnel-0.1.5.tar.gz (49kB) |████████████████████████████████| 51kB 7.6MB/s Collecting basicauth Downloading https://files.pythonhosted.org/packages/76/47/08d21ffcc837bebf3306b8295f5d179f9bc498f6235ebf4a4e38be57839c/basicauth-0.4.1-py2.py3-none-any.whl Collecting tabulator[cchardet]>=1.42.0 Downloading https://files.pythonhosted.org/packages/01/94/7669fdcedd2c113fb942f3232d8c35af1a94d2bfea03242a1f15e61cd44c/tabulator-1.49.2-py2.py3-none-any.whl (70kB) |████████████████████████████████| 71kB 9.3MB/s Requirement already satisfied: beautifulsoup4 in /usr/local/lib/python3.6/dist-packages (from hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo==0.1.0.dev0) (4.6.3) Collecting pyaml Downloading https://files.pythonhosted.org/packages/15/c4/1310a054d33abc318426a956e7d6df0df76a6ddfa9c66f6310274fb75d42/pyaml-20.4.0-py2.py3-none-any.whl Collecting email-validator Downloading https://files.pythonhosted.org/packages/8b/f5/26dc56e8e5b3441e766c8c359be9a28d2355902ab8b2140a2d5988da675e/email_validator-1.1.1-py2.py3-none-any.whl Requirement already satisfied: html5lib in /usr/local/lib/python3.6/dist-packages (from hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo==0.1.0.dev0) (1.0.1) Collecting colorlog Downloading https://files.pythonhosted.org/packages/00/0d/22c73c2eccb21dd3498df7d22c0b1d4a30f5a5fb3feb64e1ce06bc247747/colorlog-4.1.0-py2.py3-none-any.whl Requirement already satisfied: pyparsing>=2.0.2 in /usr/local/lib/python3.6/dist-packages (from packaging->bleach->nbconvert->jupyter->task-geo==0.1.0.dev0) (2.4.7) Requirement already satisfied: pycparser in /usr/local/lib/python3.6/dist-packages (from cffi!=1.11.3,>=1.8->cryptography>=2.8->pyOpenSSL->hdx-python-api->task-geo==0.1.0.dev0) (2.20) Requirement already satisfied: pyyaml in /usr/local/lib/python3.6/dist-packages (from yamlloader->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo==0.1.0.dev0) (3.13) Collecting paramiko>=1.15.2 Downloading https://files.pythonhosted.org/packages/06/1e/1e08baaaf6c3d3df1459fd85f0e7d2d6aa916f33958f151ee1ecc9800971/paramiko-2.7.1-py2.py3-none-any.whl (206kB) |████████████████████████████████| 215kB 53.0MB/s Requirement already satisfied: click>=6.0 in /usr/local/lib/python3.6/dist-packages (from tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo==0.1.0.dev0) (7.1.2) Collecting linear-tsv>=1.0 Downloading https://files.pythonhosted.org/packages/82/e5/03207a0f11e1d60df85b97b61704ed701b725a7c2feaf83f7bfbd0c2d83e/linear-tsv-1.1.0.tar.gz Collecting ijson>=3.0.3 Downloading https://files.pythonhosted.org/packages/11/82/03c325c85196744658c6d095c1e90dbd408595c596fc136b2157b2edaa10/ijson-3.0.4-cp36-cp36m-manylinux1_x86_64.whl (98kB) |████████████████████████████████| 102kB 12.2MB/s Collecting unicodecsv>=0.14 Downloading https://files.pythonhosted.org/packages/6f/a4/691ab63b17505a26096608cc309960b5a6bdf39e4ba1a793d5f9b1a53270/unicodecsv-0.14.1.tar.gz Collecting jsonlines>=1.1 Downloading https://files.pythonhosted.org/packages/4f/9a/ab96291470e305504aa4b7a2e0ec132e930da89eb3ca7a82fbe03167c131/jsonlines-1.2.0-py2.py3-none-any.whl Collecting openpyxl>=2.6 Downloading https://files.pythonhosted.org/packages/95/8c/83563c60489954e5b80f9e2596b93a68e1ac4e4a730deb1aae632066d704/openpyxl-3.0.3.tar.gz (172kB) |████████████████████████████████| 174kB 44.7MB/s Requirement already satisfied: boto3>=1.9 in /usr/local/lib/python3.6/dist-packages (from tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo==0.1.0.dev0) (1.13.13) Requirement already satisfied: sqlalchemy>=0.9.6 in /usr/local/lib/python3.6/dist-packages (from tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo==0.1.0.dev0) (1.3.17) Collecting cchardet>=2.0; extra == "cchardet" Downloading https://files.pythonhosted.org/packages/1e/c5/7e1a0d7b4afd83d6f8de794fce82820ec4c5136c6d52e14000822681a842/cchardet-2.1.6-cp36-cp36m-manylinux2010_x86_64.whl (241kB) |████████████████████████████████| 245kB 49.9MB/s Collecting dnspython>=1.15.0 Downloading https://files.pythonhosted.org/packages/ec/d3/3aa0e7213ef72b8585747aa0e271a9523e713813b9a20177ebe1e939deb0/dnspython-1.16.0-py2.py3-none-any.whl (188kB) |████████████████████████████████| 194kB 50.3MB/s Collecting pynacl>=1.0.1 Downloading https://files.pythonhosted.org/packages/9d/57/2f5e6226a674b2bcb6db531e8b383079b678df5b10cdaa610d6cf20d77ba/PyNaCl-1.4.0-cp35-abi3-manylinux1_x86_64.whl (961kB) |████████████████████████████████| 962kB 43.8MB/s Collecting bcrypt>=3.1.3 Downloading https://files.pythonhosted.org/packages/8b/1d/82826443777dd4a624e38a08957b975e75df859b381ae302cfd7a30783ed/bcrypt-3.1.7-cp34-abi3-manylinux1_x86_64.whl (56kB) |████████████████████████████████| 61kB 8.9MB/s Requirement already satisfied: jdcal in /usr/local/lib/python3.6/dist-packages (from openpyxl>=2.6->tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo==0.1.0.dev0) (1.4.1) Requirement already satisfied: et_xmlfile in /usr/local/lib/python3.6/dist-packages (from openpyxl>=2.6->tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo==0.1.0.dev0) (1.0.1) Requirement already satisfied: s3transfer<0.4.0,>=0.3.0 in /usr/local/lib/python3.6/dist-packages (from boto3>=1.9->tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo==0.1.0.dev0) (0.3.3) Requirement already satisfied: botocore<1.17.0,>=1.16.13 in /usr/local/lib/python3.6/dist-packages (from boto3>=1.9->tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo==0.1.0.dev0) (1.16.13) Requirement already satisfied: jmespath<1.0.0,>=0.7.1 in /usr/local/lib/python3.6/dist-packages (from boto3>=1.9->tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo==0.1.0.dev0) (0.10.0) Requirement already satisfied: docutils<0.16,>=0.10 in /usr/local/lib/python3.6/dist-packages (from botocore<1.17.0,>=1.16.13->boto3>=1.9->tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo==0.1.0.dev0) (0.15.2) Building wheels for collected packages: task-geo, ckanapi, libhxl, python-io-wrapper, jsonpath-ng, ratelimit, sshtunnel, linear-tsv, unicodecsv, openpyxl Building wheel for task-geo (setup.py) ... done Created wheel for task-geo: filename=task_geo-0.1.0.dev0-py2.py3-none-any.whl size=177679 sha256=3aa893dbe5860d1ce720aeded4caff21d60f0266716c8b8aa18f73fb57caf310 Stored in directory: /tmp/pip-ephem-wheel-cache-vnqx9at_/wheels/4f/c4/a4/33f04d80a745ae3ff088cae6f8cda3b46c48e007585d9ff0bf Building wheel for ckanapi (setup.py) ... done Created wheel for ckanapi: filename=ckanapi-4.3-cp36-none-any.whl size=38647 sha256=efe395add978afcfd1aecb4bd03a9221f1219d2200748ceaf75f52d4979ba87d Stored in directory: /root/.cache/pip/wheels/41/f2/fb/c8ce857007de64cc6b36b8f1048272396bc0817c35ee3a3e73 Building wheel for libhxl (setup.py) ... done Created wheel for libhxl: filename=libhxl-4.20-cp36-none-any.whl size=85431 sha256=0c1b5d43e55bab87c3557292e9f767660793c9f0aa66cb58ef508658cdb3480a Stored in directory: /root/.cache/pip/wheels/31/9a/17/9ff6d21ecc021de390dad3cb7972acaddb305580e1ff8e1a2e Building wheel for python-io-wrapper (setup.py) ... done Created wheel for python-io-wrapper: filename=python_io_wrapper-0.1-cp36-none-any.whl size=2490 sha256=2466931813d866c03be477a346f6af1eb6cb8ea24298a2485cac37c4a54ef4d8 Stored in directory: /root/.cache/pip/wheels/6b/26/be/da3c0a774901c557a0bee985e7aade5b9db75fe4dc8ef99ced Building wheel for jsonpath-ng (setup.py) ... done Created wheel for jsonpath-ng: filename=jsonpath_ng-1.5.1-cp36-none-any.whl size=23919 sha256=31152cad940087e3fd2ea48b252fe47269193c4b4fa42357e6ea94d01e08d3a2 Stored in directory: /root/.cache/pip/wheels/22/e6/94/298d44da8cc99fca216343afc8d877348146d583beac99ec49 Building wheel for ratelimit (setup.py) ... done Created wheel for ratelimit: filename=ratelimit-2.2.1-cp36-none-any.whl size=5893 sha256=7d83f65805d9f73a32d3d208dffb65401f777f71341ab59f57b5e82f95523cbb Stored in directory: /root/.cache/pip/wheels/05/d9/82/3c6044cf1a54aab9151612458446d9b17a38416869e1b1d9b8 Building wheel for sshtunnel (setup.py) ... done Created wheel for sshtunnel: filename=sshtunnel-0.1.5-py2.py3-none-any.whl size=23243 sha256=3344e2782827b30a6944ea88090c4048b4830e4719fa0bf0eb2534089b469b17 Stored in directory: /root/.cache/pip/wheels/e8/d2/38/b9791b7391f634099194ec6697fa671194f3353906d94c8f92 Building wheel for linear-tsv (setup.py) ... done Created wheel for linear-tsv: filename=linear_tsv-1.1.0-cp36-none-any.whl size=7383 sha256=a8987aa5a1fc41f0cfcd4ea4d03587ab15dd5ae633eab4aa81a787303e6b8672 Stored in directory: /root/.cache/pip/wheels/3f/8a/cb/38917fd1ef4356b9870ace7331b83417dc594bf2c029bd991f Building wheel for unicodecsv (setup.py) ... done Created wheel for unicodecsv: filename=unicodecsv-0.14.1-cp36-none-any.whl size=10768 sha256=8aade927a27069a8db5cc9d2014f381f2dca18ea897870e982d882b19ceb5130 Stored in directory: /root/.cache/pip/wheels/a6/09/e9/e800279c98a0a8c94543f3de6c8a562f60e51363ed26e71283 Building wheel for openpyxl (setup.py) ... done Created wheel for openpyxl: filename=openpyxl-3.0.3-py2.py3-none-any.whl size=241262 sha256=bc743c505ce18b8da74d160118642b3594ad02e2d4d6ec736b9928f9c492b924 Stored in directory: /root/.cache/pip/wheels/b5/85/ca/e768ac132e57e75e645a151f8badac71cc0089e7225dddf76b Successfully built task-geo ckanapi libhxl python-io-wrapper jsonpath-ng ratelimit sshtunnel linear-tsv unicodecsv openpyxl ERROR: hdx-python-utilities 2.3.4 has requirement six>=1.14.0, but you'll have six 1.12.0 which is incompatible. Installing collected packages: ckanapi, cryptography, pyOpenSSL, unidecode, python-io-wrapper, ply, jsonpath-ng, libhxl, ratelimit, psycopg2-binary, yamlloader, pynacl, bcrypt, paramiko, sshtunnel, basicauth, linear-tsv, ijson, unicodecsv, jsonlines, openpyxl, cchardet, tabulator, pyaml, dnspython, email-validator, colorlog, hdx-python-utilities, hdx-python-country, ndg-httpsclient, num2words, quantulum3, hdx-python-api, task-geo Found existing installation: openpyxl 2.5.9 Uninstalling openpyxl-2.5.9: Successfully uninstalled openpyxl-2.5.9 Successfully installed basicauth-0.4.1 bcrypt-3.1.7 cchardet-2.1.6 ckanapi-4.3 colorlog-4.1.0 cryptography-2.9.2 dnspython-1.16.0 email-validator-1.1.1 hdx-python-api-4.5.8 hdx-python-country-2.5.6 hdx-python-utilities-2.3.4 ijson-3.0.4 jsonlines-1.2.0 jsonpath-ng-1.5.1 libhxl-4.20 linear-tsv-1.1.0 ndg-httpsclient-0.5.1 num2words-0.5.10 openpyxl-3.0.3 paramiko-2.7.1 ply-3.11 psycopg2-binary-2.8.5 pyOpenSSL-19.1.0 pyaml-20.4.0 pynacl-1.4.0 python-io-wrapper-0.1 quantulum3-0.7.3 ratelimit-2.2.1 sshtunnel-0.1.5 tabulator-1.49.2 task-geo-0.1.0.dev0 unicodecsv-0.14.1 unidecode-1.1.1 yamlloader-0.5.5 Collecting git+https://github.com/coronawhy/task-ts Cloning https://github.com/coronawhy/task-ts to /tmp/pip-req-build-qk07_tgz Running command git clone -q https://github.com/coronawhy/task-ts /tmp/pip-req-build-qk07_tgz Collecting loguru Downloading https://files.pythonhosted.org/packages/80/b0/4413a201fcdcdc6789050c536d3b4ece601975ded9e0d676ef47f582348d/loguru-0.5.0-py3-none-any.whl (56kB) |████████████████████████████████| 61kB 1.2MB/s Requirement already satisfied: pandas in /usr/local/lib/python3.6/dist-packages (from corona-ts-tools==0.0.1) (1.0.3) Requirement already satisfied: tensorflow in /usr/local/lib/python3.6/dist-packages (from corona-ts-tools==0.0.1) (2.2.0) Collecting taskgeo@ git+https://github.com/CoronaWhy/task-geo.git Cloning https://github.com/CoronaWhy/task-geo.git to /tmp/pip-install-o5lx2qws/taskgeo Running command git clone -q https://github.com/CoronaWhy/task-geo.git /tmp/pip-install-o5lx2qws/taskgeo WARNING: Generating metadata for package taskgeo produced metadata for project name task-geo. Fix your #egg=taskgeo fragments. Requirement already satisfied (use --upgrade to upgrade): task-geo from git+https://github.com/CoronaWhy/task-geo.git in /usr/local/lib/python3.6/dist-packages (from corona-ts-tools==0.0.1) Requirement already satisfied: tqdm in /usr/local/lib/python3.6/dist-packages (from corona-ts-tools==0.0.1) (4.41.1) Requirement already satisfied: joblib in /usr/local/lib/python3.6/dist-packages (from corona-ts-tools==0.0.1) (0.15.1) Collecting aiocontextvars>=0.2.0; python_version < "3.7" Downloading https://files.pythonhosted.org/packages/db/c1/7a723e8d988de0a2e623927396e54b6831b68cb80dce468c945b849a9385/aiocontextvars-0.2.2-py2.py3-none-any.whl Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.6/dist-packages (from pandas->corona-ts-tools==0.0.1) (2018.9) Requirement already satisfied: python-dateutil>=2.6.1 in /usr/local/lib/python3.6/dist-packages (from pandas->corona-ts-tools==0.0.1) (2.8.1) Requirement already satisfied: numpy>=1.13.3 in /usr/local/lib/python3.6/dist-packages (from pandas->corona-ts-tools==0.0.1) (1.18.4) Requirement already satisfied: astunparse==1.6.3 in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (1.6.3) Requirement already satisfied: h5py<2.11.0,>=2.10.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (2.10.0) Requirement already satisfied: tensorboard<2.3.0,>=2.2.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (2.2.1) Requirement already satisfied: wrapt>=1.11.1 in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (1.12.1) Requirement already satisfied: grpcio>=1.8.6 in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (1.29.0) Requirement already satisfied: gast==0.3.3 in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (0.3.3) Requirement already satisfied: wheel>=0.26; python_version >= "3" in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (0.34.2) Requirement already satisfied: absl-py>=0.7.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (0.9.0) Requirement already satisfied: protobuf>=3.8.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (3.10.0) Requirement already satisfied: keras-preprocessing>=1.1.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (1.1.2) Requirement already satisfied: termcolor>=1.1.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (1.1.0) Requirement already satisfied: six>=1.12.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (1.12.0) Requirement already satisfied: opt-einsum>=2.3.2 in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (3.2.1) Requirement already satisfied: google-pasta>=0.1.8 in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (0.2.0) Requirement already satisfied: tensorflow-estimator<2.3.0,>=2.2.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (2.2.0) Requirement already satisfied: scipy==1.4.1; python_version >= "3" in /usr/local/lib/python3.6/dist-packages (from tensorflow->corona-ts-tools==0.0.1) (1.4.1) Requirement already satisfied: requests in /usr/local/lib/python3.6/dist-packages (from task-geo->corona-ts-tools==0.0.1) (2.23.0) Requirement already satisfied: jupyter in /usr/local/lib/python3.6/dist-packages (from task-geo->corona-ts-tools==0.0.1) (1.0.0) Requirement already satisfied: hdx-python-api in /usr/local/lib/python3.6/dist-packages (from task-geo->corona-ts-tools==0.0.1) (4.5.8) Collecting contextvars==2.4; python_version < "3.7" Downloading https://files.pythonhosted.org/packages/83/96/55b82d9f13763be9d672622e1b8106c85acb83edd7cc2fa5bc67cd9877e9/contextvars-2.4.tar.gz Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.3.0,>=2.2.0->tensorflow->corona-ts-tools==0.0.1) (3.2.2) Requirement already satisfied: werkzeug>=0.11.15 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.3.0,>=2.2.0->tensorflow->corona-ts-tools==0.0.1) (1.0.1) Requirement already satisfied: google-auth-oauthlib<0.5,>=0.4.1 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.3.0,>=2.2.0->tensorflow->corona-ts-tools==0.0.1) (0.4.1) Requirement already satisfied: setuptools>=41.0.0 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.3.0,>=2.2.0->tensorflow->corona-ts-tools==0.0.1) (46.4.0) Requirement already satisfied: tensorboard-plugin-wit>=1.6.0 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.3.0,>=2.2.0->tensorflow->corona-ts-tools==0.0.1) (1.6.0.post3) Requirement already satisfied: google-auth<2,>=1.6.3 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.3.0,>=2.2.0->tensorflow->corona-ts-tools==0.0.1) (1.7.2) Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests->task-geo->corona-ts-tools==0.0.1) (2.9) Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests->task-geo->corona-ts-tools==0.0.1) (2020.4.5.1) Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests->task-geo->corona-ts-tools==0.0.1) (3.0.4) Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests->task-geo->corona-ts-tools==0.0.1) (1.24.3) Requirement already satisfied: ipywidgets in /usr/local/lib/python3.6/dist-packages (from jupyter->task-geo->corona-ts-tools==0.0.1) (7.5.1) Requirement already satisfied: ipykernel in /usr/local/lib/python3.6/dist-packages (from jupyter->task-geo->corona-ts-tools==0.0.1) (4.10.1) Requirement already satisfied: qtconsole in /usr/local/lib/python3.6/dist-packages (from jupyter->task-geo->corona-ts-tools==0.0.1) (4.7.4) Requirement already satisfied: nbconvert in /usr/local/lib/python3.6/dist-packages (from jupyter->task-geo->corona-ts-tools==0.0.1) (5.6.1) Requirement already satisfied: jupyter-console in /usr/local/lib/python3.6/dist-packages (from jupyter->task-geo->corona-ts-tools==0.0.1) (5.2.0) Requirement already satisfied: notebook in /usr/local/lib/python3.6/dist-packages (from jupyter->task-geo->corona-ts-tools==0.0.1) (5.2.2) Requirement already satisfied: pyasn1 in /usr/local/lib/python3.6/dist-packages (from hdx-python-api->task-geo->corona-ts-tools==0.0.1) (0.4.8) Requirement already satisfied: pyOpenSSL in /usr/local/lib/python3.6/dist-packages (from hdx-python-api->task-geo->corona-ts-tools==0.0.1) (19.1.0) Requirement already satisfied: hdx-python-country>=2.5.6 in /usr/local/lib/python3.6/dist-packages (from hdx-python-api->task-geo->corona-ts-tools==0.0.1) (2.5.6) Requirement already satisfied: ndg-httpsclient in /usr/local/lib/python3.6/dist-packages (from hdx-python-api->task-geo->corona-ts-tools==0.0.1) (0.5.1) Requirement already satisfied: quantulum3>=0.7.3; python_version >= "3" in /usr/local/lib/python3.6/dist-packages (from hdx-python-api->task-geo->corona-ts-tools==0.0.1) (0.7.3) Requirement already satisfied: ckanapi>=4.3 in /usr/local/lib/python3.6/dist-packages (from hdx-python-api->task-geo->corona-ts-tools==0.0.1) (4.3) Collecting immutables>=0.9 Downloading https://files.pythonhosted.org/packages/99/e0/ea6fd4697120327d26773b5a84853f897a68e33d3f9376b00a8ff96e4f63/immutables-0.14-cp36-cp36m-manylinux1_x86_64.whl (98kB) |████████████████████████████████| 102kB 3.2MB/s Requirement already satisfied: importlib-metadata; python_version < "3.8" in /usr/local/lib/python3.6/dist-packages (from markdown>=2.6.8->tensorboard<2.3.0,>=2.2.0->tensorflow->corona-ts-tools==0.0.1) (1.6.0) Requirement already satisfied: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.6/dist-packages (from google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.3.0,>=2.2.0->tensorflow->corona-ts-tools==0.0.1) (1.3.0) Requirement already satisfied: cachetools<3.2,>=2.0.0 in /usr/local/lib/python3.6/dist-packages (from google-auth<2,>=1.6.3->tensorboard<2.3.0,>=2.2.0->tensorflow->corona-ts-tools==0.0.1) (3.1.1) Requirement already satisfied: rsa<4.1,>=3.1.4 in /usr/local/lib/python3.6/dist-packages (from google-auth<2,>=1.6.3->tensorboard<2.3.0,>=2.2.0->tensorflow->corona-ts-tools==0.0.1) (4.0) Requirement already satisfied: pyasn1-modules>=0.2.1 in /usr/local/lib/python3.6/dist-packages (from google-auth<2,>=1.6.3->tensorboard<2.3.0,>=2.2.0->tensorflow->corona-ts-tools==0.0.1) (0.2.8) Requirement already satisfied: traitlets>=4.3.1 in /usr/local/lib/python3.6/dist-packages (from ipywidgets->jupyter->task-geo->corona-ts-tools==0.0.1) (4.3.3) Requirement already satisfied: nbformat>=4.2.0 in /usr/local/lib/python3.6/dist-packages (from ipywidgets->jupyter->task-geo->corona-ts-tools==0.0.1) (5.0.6) Requirement already satisfied: widgetsnbextension~=3.5.0 in /usr/local/lib/python3.6/dist-packages (from ipywidgets->jupyter->task-geo->corona-ts-tools==0.0.1) (3.5.1) Requirement already satisfied: ipython>=4.0.0; python_version >= "3.3" in /usr/local/lib/python3.6/dist-packages (from ipywidgets->jupyter->task-geo->corona-ts-tools==0.0.1) (5.5.0) Requirement already satisfied: tornado>=4.0 in /usr/local/lib/python3.6/dist-packages (from ipykernel->jupyter->task-geo->corona-ts-tools==0.0.1) (4.5.3) Requirement already satisfied: jupyter-client in /usr/local/lib/python3.6/dist-packages (from ipykernel->jupyter->task-geo->corona-ts-tools==0.0.1) (5.3.4) Requirement already satisfied: qtpy in /usr/local/lib/python3.6/dist-packages (from qtconsole->jupyter->task-geo->corona-ts-tools==0.0.1) (1.9.0) Requirement already satisfied: pyzmq>=17.1 in /usr/local/lib/python3.6/dist-packages (from qtconsole->jupyter->task-geo->corona-ts-tools==0.0.1) (19.0.1) Requirement already satisfied: jupyter-core in /usr/local/lib/python3.6/dist-packages (from qtconsole->jupyter->task-geo->corona-ts-tools==0.0.1) (4.6.3) Requirement already satisfied: ipython-genutils in /usr/local/lib/python3.6/dist-packages (from qtconsole->jupyter->task-geo->corona-ts-tools==0.0.1) (0.2.0) Requirement already satisfied: pygments in /usr/local/lib/python3.6/dist-packages (from qtconsole->jupyter->task-geo->corona-ts-tools==0.0.1) (2.1.3) Requirement already satisfied: mistune<2,>=0.8.1 in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo->corona-ts-tools==0.0.1) (0.8.4) Requirement already satisfied: jinja2>=2.4 in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo->corona-ts-tools==0.0.1) (2.11.2) Requirement already satisfied: pandocfilters>=1.4.1 in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo->corona-ts-tools==0.0.1) (1.4.2) Requirement already satisfied: defusedxml in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo->corona-ts-tools==0.0.1) (0.6.0) Requirement already satisfied: testpath in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo->corona-ts-tools==0.0.1) (0.4.4) Requirement already satisfied: entrypoints>=0.2.2 in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo->corona-ts-tools==0.0.1) (0.3) Requirement already satisfied: bleach in /usr/local/lib/python3.6/dist-packages (from nbconvert->jupyter->task-geo->corona-ts-tools==0.0.1) (3.1.5) Requirement already satisfied: prompt-toolkit<2.0.0,>=1.0.0 in /usr/local/lib/python3.6/dist-packages (from jupyter-console->jupyter->task-geo->corona-ts-tools==0.0.1) (1.0.18) Requirement already satisfied: terminado>=0.3.3; sys_platform != "win32" in /usr/local/lib/python3.6/dist-packages (from notebook->jupyter->task-geo->corona-ts-tools==0.0.1) (0.8.3) Requirement already satisfied: cryptography>=2.8 in /usr/local/lib/python3.6/dist-packages (from pyOpenSSL->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (2.9.2) Requirement already satisfied: hdx-python-utilities>=2.3.4 in /usr/local/lib/python3.6/dist-packages (from hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (2.3.4) Requirement already satisfied: libhxl>=4.19; python_version >= "3" in /usr/local/lib/python3.6/dist-packages (from hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (4.20) Requirement already satisfied: inflect in /usr/local/lib/python3.6/dist-packages (from quantulum3>=0.7.3; python_version >= "3"->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (2.1.0) Requirement already satisfied: num2words in /usr/local/lib/python3.6/dist-packages (from quantulum3>=0.7.3; python_version >= "3"->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (0.5.10) Requirement already satisfied: docopt in /usr/local/lib/python3.6/dist-packages (from ckanapi>=4.3->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (0.6.2) Requirement already satisfied: python-slugify>=1.0 in /usr/local/lib/python3.6/dist-packages (from ckanapi>=4.3->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (4.0.0) Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.6/dist-packages (from importlib-metadata; python_version < "3.8"->markdown>=2.6.8->tensorboard<2.3.0,>=2.2.0->tensorflow->corona-ts-tools==0.0.1) (3.1.0) Requirement already satisfied: oauthlib>=3.0.0 in /usr/local/lib/python3.6/dist-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.3.0,>=2.2.0->tensorflow->corona-ts-tools==0.0.1) (3.1.0) Requirement already satisfied: decorator in /usr/local/lib/python3.6/dist-packages (from traitlets>=4.3.1->ipywidgets->jupyter->task-geo->corona-ts-tools==0.0.1) (4.4.2) Requirement already satisfied: jsonschema!=2.5.0,>=2.4 in /usr/local/lib/python3.6/dist-packages (from nbformat>=4.2.0->ipywidgets->jupyter->task-geo->corona-ts-tools==0.0.1) (2.6.0) Requirement already satisfied: pickleshare in /usr/local/lib/python3.6/dist-packages (from ipython>=4.0.0; python_version >= "3.3"->ipywidgets->jupyter->task-geo->corona-ts-tools==0.0.1) (0.7.5) Requirement already satisfied: simplegeneric>0.8 in /usr/local/lib/python3.6/dist-packages (from ipython>=4.0.0; python_version >= "3.3"->ipywidgets->jupyter->task-geo->corona-ts-tools==0.0.1) (0.8.1) Requirement already satisfied: pexpect; sys_platform != "win32" in /usr/local/lib/python3.6/dist-packages (from ipython>=4.0.0; python_version >= "3.3"->ipywidgets->jupyter->task-geo->corona-ts-tools==0.0.1) (4.8.0) Requirement already satisfied: MarkupSafe>=0.23 in /usr/local/lib/python3.6/dist-packages (from jinja2>=2.4->nbconvert->jupyter->task-geo->corona-ts-tools==0.0.1) (1.1.1) Requirement already satisfied: packaging in /usr/local/lib/python3.6/dist-packages (from bleach->nbconvert->jupyter->task-geo->corona-ts-tools==0.0.1) (20.4) Requirement already satisfied: webencodings in /usr/local/lib/python3.6/dist-packages (from bleach->nbconvert->jupyter->task-geo->corona-ts-tools==0.0.1) (0.5.1) Requirement already satisfied: wcwidth in /usr/local/lib/python3.6/dist-packages (from prompt-toolkit<2.0.0,>=1.0.0->jupyter-console->jupyter->task-geo->corona-ts-tools==0.0.1) (0.1.9) Requirement already satisfied: ptyprocess; os_name != "nt" in /usr/local/lib/python3.6/dist-packages (from terminado>=0.3.3; sys_platform != "win32"->notebook->jupyter->task-geo->corona-ts-tools==0.0.1) (0.6.0) Requirement already satisfied: cffi!=1.11.3,>=1.8 in /usr/local/lib/python3.6/dist-packages (from cryptography>=2.8->pyOpenSSL->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.14.0) Requirement already satisfied: sshtunnel in /usr/local/lib/python3.6/dist-packages (from hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (0.1.5) Requirement already satisfied: colorlog in /usr/local/lib/python3.6/dist-packages (from hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (4.1.0) Requirement already satisfied: beautifulsoup4 in /usr/local/lib/python3.6/dist-packages (from hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (4.6.3) Requirement already satisfied: tabulator[cchardet]>=1.42.0 in /usr/local/lib/python3.6/dist-packages (from hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.49.2) Requirement already satisfied: ratelimit in /usr/local/lib/python3.6/dist-packages (from hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (2.2.1) Requirement already satisfied: basicauth in /usr/local/lib/python3.6/dist-packages (from hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (0.4.1) Requirement already satisfied: email-validator in /usr/local/lib/python3.6/dist-packages (from hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.1.1) Requirement already satisfied: pyaml in /usr/local/lib/python3.6/dist-packages (from hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (20.4.0) Requirement already satisfied: html5lib in /usr/local/lib/python3.6/dist-packages (from hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.0.1) Requirement already satisfied: yamlloader in /usr/local/lib/python3.6/dist-packages (from hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (0.5.5) Requirement already satisfied: psycopg2-binary in /usr/local/lib/python3.6/dist-packages (from hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (2.8.5) Requirement already satisfied: ply in /usr/local/lib/python3.6/dist-packages (from libhxl>=4.19; python_version >= "3"->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (3.11) Requirement already satisfied: unidecode in /usr/local/lib/python3.6/dist-packages (from libhxl>=4.19; python_version >= "3"->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.1.1) Requirement already satisfied: python-io-wrapper in /usr/local/lib/python3.6/dist-packages (from libhxl>=4.19; python_version >= "3"->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (0.1) Requirement already satisfied: jsonpath-ng in /usr/local/lib/python3.6/dist-packages (from libhxl>=4.19; python_version >= "3"->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.5.1) Requirement already satisfied: xlrd in /usr/local/lib/python3.6/dist-packages (from libhxl>=4.19; python_version >= "3"->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.1.0) Requirement already satisfied: text-unidecode>=1.3 in /usr/local/lib/python3.6/dist-packages (from python-slugify>=1.0->ckanapi>=4.3->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.3) Requirement already satisfied: pyparsing>=2.0.2 in /usr/local/lib/python3.6/dist-packages (from packaging->bleach->nbconvert->jupyter->task-geo->corona-ts-tools==0.0.1) (2.4.7) Requirement already satisfied: pycparser in /usr/local/lib/python3.6/dist-packages (from cffi!=1.11.3,>=1.8->cryptography>=2.8->pyOpenSSL->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (2.20) Requirement already satisfied: paramiko>=1.15.2 in /usr/local/lib/python3.6/dist-packages (from sshtunnel->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (2.7.1) Requirement already satisfied: jsonlines>=1.1 in /usr/local/lib/python3.6/dist-packages (from tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.2.0) Requirement already satisfied: unicodecsv>=0.14 in /usr/local/lib/python3.6/dist-packages (from tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (0.14.1) Requirement already satisfied: openpyxl>=2.6 in /usr/local/lib/python3.6/dist-packages (from tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (3.0.3) Requirement already satisfied: sqlalchemy>=0.9.6 in /usr/local/lib/python3.6/dist-packages (from tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.3.17) Requirement already satisfied: ijson>=3.0.3 in /usr/local/lib/python3.6/dist-packages (from tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (3.0.4) Requirement already satisfied: linear-tsv>=1.0 in /usr/local/lib/python3.6/dist-packages (from tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.1.0) Requirement already satisfied: click>=6.0 in /usr/local/lib/python3.6/dist-packages (from tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (7.1.2) Requirement already satisfied: boto3>=1.9 in /usr/local/lib/python3.6/dist-packages (from tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.13.13) Requirement already satisfied: cchardet>=2.0; extra == "cchardet" in /usr/local/lib/python3.6/dist-packages (from tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (2.1.6) Requirement already satisfied: dnspython>=1.15.0 in /usr/local/lib/python3.6/dist-packages (from email-validator->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.16.0) Requirement already satisfied: PyYAML in /usr/local/lib/python3.6/dist-packages (from pyaml->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (3.13) Requirement already satisfied: bcrypt>=3.1.3 in /usr/local/lib/python3.6/dist-packages (from paramiko>=1.15.2->sshtunnel->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (3.1.7) Requirement already satisfied: pynacl>=1.0.1 in /usr/local/lib/python3.6/dist-packages (from paramiko>=1.15.2->sshtunnel->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.4.0) Requirement already satisfied: jdcal in /usr/local/lib/python3.6/dist-packages (from openpyxl>=2.6->tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.4.1) Requirement already satisfied: et-xmlfile in /usr/local/lib/python3.6/dist-packages (from openpyxl>=2.6->tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.0.1) Requirement already satisfied: s3transfer<0.4.0,>=0.3.0 in /usr/local/lib/python3.6/dist-packages (from boto3>=1.9->tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (0.3.3) Requirement already satisfied: botocore<1.17.0,>=1.16.13 in /usr/local/lib/python3.6/dist-packages (from boto3>=1.9->tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (1.16.13) Requirement already satisfied: jmespath<1.0.0,>=0.7.1 in /usr/local/lib/python3.6/dist-packages (from boto3>=1.9->tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (0.10.0) Requirement already satisfied: docutils<0.16,>=0.10 in /usr/local/lib/python3.6/dist-packages (from botocore<1.17.0,>=1.16.13->boto3>=1.9->tabulator[cchardet]>=1.42.0->hdx-python-utilities>=2.3.4->hdx-python-country>=2.5.6->hdx-python-api->task-geo->corona-ts-tools==0.0.1) (0.15.2) Building wheels for collected packages: corona-ts-tools, task-geo, task-geo, contextvars Building wheel for corona-ts-tools (setup.py) ... done Created wheel for corona-ts-tools: filename=corona_ts_tools-0.0.1-cp36-none-any.whl size=9369 sha256=7f1ce7611fd403214db6b7807e01c287024d28d7a91c52950df38838e684b5f5 Stored in directory: /tmp/pip-ephem-wheel-cache-7tmo09xq/wheels/84/8c/21/775765ae58f390facc7a50b890c997befead02d4a4e4871a8e Building wheel for task-geo (setup.py) ... done Created wheel for task-geo: filename=task_geo-0.1.0.dev0-py2.py3-none-any.whl size=177679 sha256=7b55c117095197c6d0e375af32ffb4d66e3e0f439360936547c158e5307fb0fe Stored in directory: /tmp/pip-ephem-wheel-cache-7tmo09xq/wheels/4f/c4/a4/33f04d80a745ae3ff088cae6f8cda3b46c48e007585d9ff0bf Building wheel for task-geo (setup.py) ... error ERROR: Failed building wheel for task-geo Running setup.py clean for task-geo ERROR: Failed cleaning build dir for task-geo Building wheel for contextvars (setup.py) ... done Created wheel for contextvars: filename=contextvars-2.4-cp36-none-any.whl size=7666 sha256=ebf80f5fb3ab35cfcfe717a8660d20791060447c69a1d0dd7cd10d9af0099558 Stored in directory: /root/.cache/pip/wheels/a5/7d/68/1ebae2668bda2228686e3c1cf16f2c2384cea6e9334ad5f6de Successfully built corona-ts-tools task-geo contextvars Failed to build task-geo ERROR: corona-ts-tools 0.0.1 requires taskgeo@ git+https://github.com/CoronaWhy/task-geo.git, which is not installed. Installing collected packages: immutables, contextvars, aiocontextvars, loguru, corona-ts-tools Successfully installed aiocontextvars-0.2.2 contextvars-2.4 corona-ts-tools-0.0.1 immutables-0.14 loguru-0.5.0
Now that we have all the basic things setup lets define our config file. Since mobility data has a longer lag we will need to define our parameter sweeps to have a longer lookback window.
def make_config_file(file_path, df_len, weight_path=None):
run = wandb.init(project="covid_forecast", entity="covid")
wandb_config = wandb.config
train_number = df_len * .7
validation_number = df_len *.9
config_default={
"model_name": "CustomTransformerDecoder",
"model_type": "PyTorch",
"model_params": {
"seq_length":wandb_config["forecast_history"],
"n_time_series":9,
"output_seq_length":wandb_config["out_seq_length"],
"n_layers_encoder": wandb_config["number_encoder_layers"],
"use_mask": wandb_config["use_mask"]
},
"dataset_params":
{ "class": "default",
"training_path": file_path,
"validation_path": file_path,
"test_path": file_path,
"batch_size":wandb_config["batch_size"],
"forecast_history":wandb_config["forecast_history"],
"forecast_length":wandb_config["out_seq_length"],
"train_end": int(train_number),
"valid_start":int(train_number+1),
"valid_end": int(validation_number),
"target_col": ["new_cases"],
"relevant_cols": ["new_cases", "month", "weekday", "mobility_retail_recreation", "mobility_grocery_pharmacy", "mobility_parks", "mobility_transit_stations", "mobility_workplaces", "mobility_residential"],
"scaler": "StandardScaler",
"interpolate": False
},
"training_params":
{
"criterion":"MSE",
"optimizer": "Adam",
"optim_params":
{
},
"lr": wandb_config["lr"],
"epochs": 10,
"batch_size":wandb_config["batch_size"]
},
"GCS": True,
"early_stopping":
{
"patience":3
},
"sweep":True,
"wandb":False,
"forward_params":{},
"metrics":["MSE"],
"inference_params":
{
"datetime_start":"2020-04-21",
"hours_to_forecast":10,
"test_csv_path":file_path,
"decoder_params":{
"decoder_function": "simple_decode",
"unsqueeze_dim": 1
},
"dataset_params":{
"file_path": file_path,
"forecast_history":wandb_config["forecast_history"],
"forecast_length":wandb_config["out_seq_length"],
"relevant_cols": ["new_cases", "month", "weekday", "mobility_retail_recreation", "mobility_grocery_pharmacy", "mobility_parks", "mobility_transit_stations", "mobility_workplaces", "mobility_residential"],
"target_col": ["new_cases"],
"scaling": "StandardScaler",
"interpolate_param": False
}
},
"weight_path_add":{
"excluded_layers":["out_length_lay.weight", "out_length_lay.bias", "dense_shape.weight", "dense_shape.bias"]
}
}
if weight_path:
config_default["weight_path"] = weight_path
wandb.config.update(config_default)
return config_default
wandb_sweep_config_full = {
"name": "Default sweep",
"method": "grid",
"parameters": {
"batch_size": {
"values": [2, 5, 10, 20]
},
"lr":{
"values":[0.001, 0.002, 0.0001, .01]
},
"forecast_history":{
"values":[10, 11, 15]
},
"out_seq_length":{
"values":[1, 2, 3, 4, 5]
},
"number_encoder_layers":
{
"values":[1, 2, 3]
},
"use_mask":{
"values":[True, False]
}
}
}
os.environ['MODEL_BUCKET'] = "coronaviruspublicdata"
os.environ["ENVIRONMENT_GCP"] = "Colab"
os.environ["GCP_PROJECT"] = "gmap-997"
!gsutil cp gs://predict_cfs/experiments/25_May_202010_29PM_model.pth .
!gsutil cp gs://coronaviruspublicdata/experiments/26_May_202012_59AM_model.pth .
Copying gs://predict_cfs/experiments/25_May_202010_29PM_model.pth... / [1 files][ 4.7 MiB/ 4.7 MiB] Operation completed over 1 objects/4.7 MiB. Copying gs://coronaviruspublicdata/experiments/26_May_202012_59AM_model.pth... / [1 files][ 2.7 MiB/ 2.7 MiB] Operation completed over 1 objects/2.7 MiB.
import glob
from corona_ts.data_utils.data_crawler import load_data
from corona_ts.data_utils.data_creator import loop_through_locations
from corona_ts.data_utils.data_creator import region_df_format
!mkdir dir /usr/local/lib/python3.6/dist-packages/data
df = load_data()
df['full_county'] = df['region'] + "_" + df['sub_region']
important_cities_list = ["United_States__California__Los_Angeles_County", "United_States__Illinois__Cook_County", "United_States__Arizona__Maricopa_County", "United_States__Massachusetts__Middlesex_County", "United_States__Texas__Dallas_County", "United_States__Texas__Harris_County", "United_States__Florida__Miami Dade_County", "United_States__California__Riverside_County", "United_States__Colorado__Denver_County", "United_States__Ohio__Cuyahoga_County", "United_States__New York__Queens_County", "United_States__New York__Bronx_County"]
mkdir: cannot create directory ‘dir’: File exists mkdir: cannot create directory ‘/usr/local/lib/python3.6/dist-packages/data’: File exists
/usr/local/lib/python3.6/dist-packages/IPython/core/interactiveshell.py:2882: DtypeWarning: Columns (2) have mixed types.Specify dtype option on import or set low_memory=False. exec(code_obj, self.user_global_ns, self.user_ns) /usr/local/lib/python3.6/dist-packages/corona_ts/data_utils/data_crawler.py:150: DtypeWarning: Columns (3) have mixed types.Specify dtype option on import or set low_memory=False. mobility_df = fetch_mobility_data() 2020-05-28 17:04:03.821 | INFO | corona_ts.data_utils.data_crawler:_treat_mobility_missing_values:121 - Treat mobility missing values.
#df_list = loop_through_locations(df)
df['country'] = df['country'].str.replace(" ","_")
df['sub_region'] = df['country'] +"__"+df['region'].str.replace(" ", "_") + "__"+df['sub_region'].str.replace(" ", "_")
def loop_special_counties(special_counties_list):
for county in special_counties_list:
region = region_df_format(df,county)
file_path, len_df, file_path2 = format_corona_data(region, county)
sweep_id = wandb.sweep(wandb_sweep_config_full,'covid', 'covid_forecast')
wandb.agent(sweep_id, lambda: train_function("PyTorch", make_config_file(file_path2, len(region), weight_path="25_May_202010_29PM_model.pth")))
loop_special_counties(important_cities_list)
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:46: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
level country region ... weekday month datetime date ... 2020-02-15 sub_region United_States California ... 5 2 2020-02-15 2020-02-16 sub_region United_States California ... 6 2 2020-02-16 2020-02-17 sub_region United_States California ... 0 2 2020-02-17 2020-02-18 sub_region United_States California ... 1 2 2020-02-18 2020-02-19 sub_region United_States California ... 2 2 2020-02-19 2020-02-20 sub_region United_States California ... 3 2 2020-02-20 2020-02-21 sub_region United_States California ... 4 2 2020-02-21 2020-02-22 sub_region United_States California ... 5 2 2020-02-22 2020-02-23 sub_region United_States California ... 6 2 2020-02-23 [9 rows x 25 columns] Create sweep with ID: 65vo8egg Sweep URL: https://app.wandb.ai/covid/covid_forecast/sweeps/65vo8egg
INFO:wandb.wandb_agent:Running runs: [] INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ln8h14nh with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 1 use_mask: True wandb: Agent Started Run: ln8h14nh
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ln8h14nh INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmxuOGgxNG5oOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/media/graph/graph_0_summary_eea37da3.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/media
The running loss is: 15.545928395004012 The number of items in train is: 28 The loss for epoch 0 0.5552117283930004 The running loss is: 21.661523299873807 The number of items in train is: 28 The loss for epoch 1 0.7736258321383502 The running loss is: 14.07579830638133 The number of items in train is: 28 The loss for epoch 2 0.5027070823707618
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-history.jsonl
The running loss is: 11.572435699170455 The number of items in train is: 28 The loss for epoch 3 0.4133012749703734 1 The running loss is: 11.783770642941818 The number of items in train is: 28 The loss for epoch 4 0.42084895153363633
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-history.jsonl
The running loss is: 9.61011728970334 The number of items in train is: 28 The loss for epoch 5 0.3432184746322621 1 The running loss is: 13.171849727863446 The number of items in train is: 28 The loss for epoch 6 0.4704232045665516 The running loss is: 9.562567624612711 The number of items in train is: 28 The loss for epoch 7 0.34152027230759685 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-history.jsonl
The running loss is: 11.004484111152124 The number of items in train is: 28 The loss for epoch 8 0.3930172896840044 2 The running loss is: 10.858391232584836 The number of items in train is: 28 The loss for epoch 9 0.38779968687802985 Data saved to: 28_May_202005_07PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['ln8h14nh']
Data saved to: 28_May_202005_07PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 856.106934
66 2020-04-22 sub_region ... 66 898.349731
67 2020-04-23 sub_region ... 67 857.475891
68 2020-04-24 sub_region ... 68 861.468872
69 2020-04-25 sub_region ... 69 885.428162
70 2020-04-26 sub_region ... 70 897.812500
71 2020-04-27 sub_region ... 71 894.293274
72 2020-04-28 sub_region ... 72 929.067383
73 2020-04-29 sub_region ... 73 878.263550
74 2020-04-30 sub_region ... 74 869.549072
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ln8h14nh
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/media/plotly/test_plot_20_314de4b9.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170731-ln8h14nh/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ln8h14nh INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: n2zvztni with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 1 use_mask: False wandb: Agent Started Run: n2zvztni
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/n2zvztni INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm4yenZ6dG5pOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
15.36638443195261
The number of items in train is:
28
The loss for epoch 0
0.5487994439983075
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/media/graph/graph_0_summary_fe13b12e.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/media
The running loss is: 19.380546062253416 The number of items in train is: 28 The loss for epoch 1 0.6921623593661934 The running loss is: 13.36081693135202 The number of items in train is: 28 The loss for epoch 2 0.4771720332625721 The running loss is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-history.jsonl
11.953934513032436 The number of items in train is: 28 The loss for epoch 3 0.4269262326083013 1 The running loss is: 12.163511937658768 The number of items in train is: 28 The loss for epoch 4 0.43441114063067027 The running loss is: 9.378338906506542 The number of items in train is: 28 The loss for epoch 5 0.33494067523237653 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-history.jsonl
The running loss is: 12.704894061491359 The number of items in train is: 28 The loss for epoch 6 0.4537462164818343 The running loss is: 11.437989893951453 The number of items in train is: 28 The loss for epoch 7 0.4084996390696948 1 The running loss is: 11.367679933318868 The number of items in train is: 28 The loss for epoch 8 0.40598856904710245
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-history.jsonl
The running loss is: 11.42585400538519 The number of items in train is: 28 The loss for epoch 9 0.40806621447804253 1 Data saved to: 28_May_202005_07PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_07PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['n2zvztni'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 802.753906
66 2020-04-22 sub_region ... 66 877.084839
67 2020-04-23 sub_region ... 67 830.681274
68 2020-04-24 sub_region ... 68 816.030273
69 2020-04-25 sub_region ... 69 821.921509
70 2020-04-26 sub_region ... 70 826.182373
71 2020-04-27 sub_region ... 71 846.830322
72 2020-04-28 sub_region ... 72 856.710266
73 2020-04-29 sub_region ... 73 857.412720
74 2020-04-30 sub_region ... 74 841.012085
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/media/plotly/test_plot_20_e30096eb.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: n2zvztni
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170746-n2zvztni/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: n2zvztni INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: l1s6ccr7 with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 2 use_mask: True wandb: Agent Started Run: l1s6ccr7
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/l1s6ccr7 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmwxczZjY3I3OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
10.473457891494036
The number of items in train is:
27
The loss for epoch 0
0.38790584783311244
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/media/graph/graph_0_summary_081c54c1.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/media
The running loss is: 12.50141921825707 The number of items in train is: 27 The loss for epoch 1 0.4630155266021137 1 The running loss is: 12.783501834142953 The number of items in train is: 27 The loss for epoch 2 0.47346303089418346
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-summary.json
The running loss is: 5.422599470242858 The number of items in train is: 27 The loss for epoch 3 0.20083701741640214 1 The running loss is: 5.399664411786944 The number of items in train is: 27 The loss for epoch 4 0.19998757080692384 The running loss is: 4.413807349279523 The number of items in train is: 27 The loss for epoch 5 0.16347434626961196
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-summary.json
The running loss is: 4.564219214487821 The number of items in train is: 27 The loss for epoch 6 0.16904515609214152 The running loss is: 4.552914118394256 The number of items in train is: 27 The loss for epoch 7 0.16862644882941688 1 The running loss is: 4.2418790506199 The number of items in train is: 27 The loss for epoch 8 0.15710663150444074 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-summary.json
The running loss is: 4.331287162844092 The number of items in train is: 27 The loss for epoch 9 0.16041804306829968 Data saved to: 28_May_202005_08PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_08PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['l1s6ccr7']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 839.469238
66 2020-04-22 sub_region ... 66 751.845520
67 2020-04-23 sub_region ... 67 796.254761
68 2020-04-24 sub_region ... 68 840.047058
69 2020-04-25 sub_region ... 69 856.757812
70 2020-04-26 sub_region ... 70 857.516724
71 2020-04-27 sub_region ... 71 805.222778
72 2020-04-28 sub_region ... 72 846.005188
73 2020-04-29 sub_region ... 73 746.322815
74 2020-04-30 sub_region ... 74 799.226501
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/media/plotly/test_plot_20_908a465b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: l1s6ccr7
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170806-l1s6ccr7/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: l1s6ccr7 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: a2slkabz with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 2 use_mask: False wandb: Agent Started Run: a2slkabz
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/a2slkabz INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-metadata.json INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmEyc2xrYWJ6OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/media/graph/graph_0_summary_cb39fdfd.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/media
The running loss is: 10.56319503299892 The number of items in train is: 27 The loss for epoch 0 0.39122944566662665 The running loss is: 12.365171766839921 The number of items in train is: 27 The loss for epoch 1 0.45796932469777485 1 The running loss is: 10.063402196858078 The number of items in train is: 27 The loss for epoch 2 0.37271859988363254
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-summary.json
The running loss is: 5.191733076237142 The number of items in train is: 27 The loss for epoch 3 0.19228641023100526 1 The running loss is: 5.048446481116116 The number of items in train is: 27 The loss for epoch 4 0.1869794993005969
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-summary.json
The running loss is: 4.6334264650940895 The number of items in train is: 27 The loss for epoch 5 0.17160838759607738 1 The running loss is: 4.809368921909481 The number of items in train is: 27 The loss for epoch 6 0.17812477488553635 The running loss is: 4.469904407858849 The number of items in train is: 27 The loss for epoch 7 0.16555201510588327
INFO:wandb.wandb_agent:Running runs: ['a2slkabz']
1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-summary.json
The running loss is: 4.4312631585635245 The number of items in train is: 27 The loss for epoch 8 0.164120857724575 The running loss is: 4.365964470198378 The number of items in train is: 27 The loss for epoch 9 0.1617023877851251 1 Data saved to: 28_May_202005_08PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_08PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 820.661011
66 2020-04-22 sub_region ... 66 731.493591
67 2020-04-23 sub_region ... 67 777.978394
68 2020-04-24 sub_region ... 68 817.230530
69 2020-04-25 sub_region ... 69 833.734131
70 2020-04-26 sub_region ... 70 830.960205
71 2020-04-27 sub_region ... 71 781.981934
72 2020-04-28 sub_region ... 72 821.894836
73 2020-04-29 sub_region ... 73 732.720215
74 2020-04-30 sub_region ... 74 782.010132
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/media/plotly/test_plot_20_a6760719.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: a2slkabz
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170816-a2slkabz/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: a2slkabz INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: jgun01i3 with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 3 use_mask: True wandb: Agent Started Run: jgun01i3
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/jgun01i3 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmpndW4wMWkzOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
22.41447990387678
The number of items in train is:
27
The loss for epoch 0
0.8301659223658068
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/media/graph/graph_0_summary_7ebd0d1d.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/media
The running loss is: 20.35543255507946 The number of items in train is: 27 The loss for epoch 1 0.7539049094473874 1 The running loss is: 19.27651271224022 The number of items in train is: 27 The loss for epoch 2 0.7139449152681563 The running loss is: 12.351722501218319 The number of items in train is: 27 The loss for epoch 3 0.4574712037488266
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-history.jsonl
1 The running loss is: 9.813126027584076 The number of items in train is: 27 The loss for epoch 4 0.36344911213274356 The running loss is: 6.200845863670111 The number of items in train is: 27 The loss for epoch 5 0.2296609579137078 1 The running loss is: 5.99026171118021 The number of items in train is: 27
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-summary.json
The loss for epoch 6 0.2218615448585263
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-history.jsonl
The running loss is: 5.51112406142056 The number of items in train is: 27 The loss for epoch 7 0.20411570597853926 1 The running loss is: 5.962472440674901 The number of items in train is: 27 The loss for epoch 8 0.22083231261758893
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-history.jsonl
The running loss is: 5.207641955465078 The number of items in train is: 27 The loss for epoch 9 0.1928756279801881 1 Data saved to: 28_May_202005_08PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_08PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['jgun01i3'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 731.740784
66 2020-04-22 sub_region ... 66 738.606201
67 2020-04-23 sub_region ... 67 737.777405
68 2020-04-24 sub_region ... 68 726.210022
69 2020-04-25 sub_region ... 69 719.908691
70 2020-04-26 sub_region ... 70 719.150635
71 2020-04-27 sub_region ... 71 720.189941
72 2020-04-28 sub_region ... 72 731.834473
73 2020-04-29 sub_region ... 73 735.288940
74 2020-04-30 sub_region ... 74 731.156189
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/media/plotly/test_plot_20_3857be08.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: jgun01i3
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170836-jgun01i3/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: jgun01i3 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 8ont2i62 with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 3 use_mask: False wandb: Agent Started Run: 8ont2i62
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/8ont2i62 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjhvbnQyaTYyOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
22.289740599691868
The number of items in train is:
27
The loss for epoch 0
0.8255459481367359
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/media/graph/graph_0_summary_6ce8dcfc.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/media
The running loss is: 20.721937209367752 The number of items in train is: 27 The loss for epoch 1 0.7674791559025094 1 The running loss is: 19.501514554023743 The number of items in train is: 27 The loss for epoch 2 0.7222783168156942 The running loss is: 13.022847175598145 The number of items in train is: 27 The loss for epoch 3 0.4823276731703017
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-summary.json
1 The running loss is: 9.215275160968304 The number of items in train is: 27 The loss for epoch 4 0.3413064874432705 2 The running loss is: 6.453644126653671 The number of items in train is: 27 The loss for epoch 5 0.23902385654272856 The running loss is: 6.3195782247930765 The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-history.jsonl
27
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-summary.json
The loss for epoch 6 0.23405845277011395 1 The running loss is: 5.538634759373963 The number of items in train is: 27 The loss for epoch 7 0.20513462071755417 The running loss is: 6.0211376789957285 The number of items in train is: 27 The loss for epoch 8 0.22300509922206402 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-summary.json
The running loss is: 5.4065384501591325 The number of items in train is: 27 The loss for epoch 9 0.2002421648207086 Data saved to: 28_May_202005_08PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_08PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['8ont2i62'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 723.085754
66 2020-04-22 sub_region ... 66 731.862305
67 2020-04-23 sub_region ... 67 732.741089
68 2020-04-24 sub_region ... 68 723.526733
69 2020-04-25 sub_region ... 69 715.763306
70 2020-04-26 sub_region ... 70 713.154541
71 2020-04-27 sub_region ... 71 712.077576
72 2020-04-28 sub_region ... 72 725.158813
73 2020-04-29 sub_region ... 73 729.568176
74 2020-04-30 sub_region ... 74 726.045166
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/media/plotly/test_plot_20_967a32da.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 8ont2i62
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170846-8ont2i62/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 8ont2i62 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 9aousxvh with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 4 use_mask: True wandb: Agent Started Run: 9aousxvh
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/9aousxvh INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjlhb3VzeHZoOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
17.69617347419262
The number of items in train is:
26
The loss for epoch 0
0.6806220566997161
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/media/graph/graph_0_summary_412005fa.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/media/graph
The running loss is: 12.942448504269123 The number of items in train is: 26 The loss for epoch 1 0.4977864809334278 1 The running loss is: 11.007499657571316 The number of items in train is: 26 The loss for epoch 2 0.4233653714450506 2 The running loss is: 7.731042757630348 The number of items in train is: 26 The loss for epoch 3 0.297347798370398
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-summary.json
The running loss is: 4.677496463060379 The number of items in train is: 26 The loss for epoch 4 0.1799037101177069 1 The running loss is: 3.537908681668341 The number of items in train is: 26 The loss for epoch 5 0.13607341083339775 The running loss is: 3.6070269970223308 The number of items in train is: 26 The loss for epoch 6 0.13873180757778195
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-summary.json
1 The running loss is: 5.469477290287614 The number of items in train is: 26 The loss for epoch 7 0.21036451116490823 2 The running loss is: 4.373710255138576 The number of items in train is: 26 The loss for epoch 8 0.16821962519763753
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-summary.json
The running loss is: 4.385507521219552 The number of items in train is: 26 The loss for epoch 9 0.168673366200752 Data saved to: 28_May_202005_09PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_09PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['9aousxvh']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 641.062012
66 2020-04-22 sub_region ... 66 649.469788
67 2020-04-23 sub_region ... 67 662.655884
68 2020-04-24 sub_region ... 68 663.000610
69 2020-04-25 sub_region ... 69 652.813110
70 2020-04-26 sub_region ... 70 647.735291
71 2020-04-27 sub_region ... 71 634.131409
72 2020-04-28 sub_region ... 72 641.305664
73 2020-04-29 sub_region ... 73 641.875549
74 2020-04-30 sub_region ... 74 656.309570
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/media/plotly/test_plot_20_7320596b.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/media/plotly
wandb: Agent Finished Run: 9aousxvh
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170901-9aousxvh/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 9aousxvh INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: rb6loxpb with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 4 use_mask: False wandb: Agent Started Run: rb6loxpb
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/rb6loxpb INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnJiNmxveHBiOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
17.936274334788322
The number of items in train is:
26
The loss for epoch 0
0.6898567051841662
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/media/graph/graph_0_summary_30873c3d.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/media/graph
The running loss is: 12.563346169888973 The number of items in train is: 26 The loss for epoch 1 0.48320562191880667 1 The running loss is: 10.63600341975689 The number of items in train is: 26 The loss for epoch 2 0.4090770546060342 2 The running loss is: 7.032686905935407 The number of items in train is: 26 The loss for epoch 3 0.27048795792059255
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-summary.json
The running loss is: 4.1291360929608345 The number of items in train is: 26 The loss for epoch 4 0.1588129266523398 1 The running loss is: 3.5215289955958724 The number of items in train is: 26 The loss for epoch 5 0.13544342290753356 The running loss is: 4.55407845415175 The number of items in train is: 26 The loss for epoch 6 0.17515686362122113
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-summary.json
1 The running loss is: 3.83646054379642 The number of items in train is: 26 The loss for epoch 7 0.14755617476140076 2 The running loss is: 4.684140918310732 The number of items in train is: 26 The loss for epoch 8 0.18015926608887428 The running loss is: 7.284909620881081 The number of items in train is: 26 The loss for epoch 9 0.28018883157234925
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-summary.json
1 Data saved to: 28_May_202005_09PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_09PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['rb6loxpb'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 811.324219
66 2020-04-22 sub_region ... 66 801.534241
67 2020-04-23 sub_region ... 67 837.117920
68 2020-04-24 sub_region ... 68 854.320984
69 2020-04-25 sub_region ... 69 834.952698
70 2020-04-26 sub_region ... 70 831.260376
71 2020-04-27 sub_region ... 71 782.874146
72 2020-04-28 sub_region ... 72 832.120972
73 2020-04-29 sub_region ... 73 785.114258
74 2020-04-30 sub_region ... 74 817.378418
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/media/plotly/test_plot_20_d2368e1b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: rb6loxpb
INFO:wandb.run_manager:shutting down system stats and metadata service
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-events.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/media/plotly/test_plot_all_21_53c27da0.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/media/plotly
INFO:wandb.run_manager:stopping streaming files and file change observer
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170912-rb6loxpb/wandb-metadata.json
INFO:wandb.wandb_agent:Cleaning up finished run: rb6loxpb
wandb: Network error resolved after 0:00:11.417233, resuming normal operation.
INFO:wandb.wandb_agent:Agent received command: run
INFO:wandb.wandb_agent:Agent starting run with config:
batch_size: 2
forecast_history: 10
lr: 0.001
number_encoder_layers: 1
out_seq_length: 5
use_mask: False
DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: rbsc7bkb with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 1 out_seq_length: 5 use_mask: False wandb: Agent Started Run: rbsc7bkb
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/rbsc7bkb INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/wandb-metadata.json INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnJic2M3YmtiOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.wandb_agent:Running runs: ['rbsc7bkb'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/config.yaml
The running loss is: 16.377284348011017 The number of items in train is: 26 The loss for epoch 0 0.6298955518465775
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/media/graph/graph_0_summary_2fdadaf1.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/media/graph
The running loss is: 14.166790530085564 The number of items in train is: 26 The loss for epoch 1 0.5448765588494447 The running loss is: 9.428598899394274 The number of items in train is: 26 The loss for epoch 2 0.3626384192074721 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/wandb-summary.json
The running loss is: 6.6330753564834595 The number of items in train is: 26 The loss for epoch 3 0.2551182829416715 2 The running loss is: 5.67338194232434 The number of items in train is: 26 The loss for epoch 4 0.2182069977817054 3 Stopping model now Data saved to: 28_May_202005_09PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_09PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 339.937469
66 2020-04-22 sub_region ... 66 348.662567
67 2020-04-23 sub_region ... 67 321.414612
68 2020-04-24 sub_region ... 68 302.783997
69 2020-04-25 sub_region ... 69 324.099854
70 2020-04-26 sub_region ... 70 317.279297
71 2020-04-27 sub_region ... 71 371.108948
72 2020-04-28 sub_region ... 72 297.625122
73 2020-04-29 sub_region ... 73 384.647125
74 2020-04-30 sub_region ... 74 332.773010
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: rbsc7bkb
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/media/plotly/test_plot_all_11_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/media/plotly/test_plot_10_807ad6e3.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170933-rbsc7bkb/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: rbsc7bkb INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: tc6k32xx with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 1 use_mask: True wandb: Agent Started Run: tc6k32xx
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/tc6k32xx INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnRjNmszMnh4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/media/graph/graph_0_summary_cdca1200.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/media/graph
The running loss is: 16.114518700167537 The number of items in train is: 28 The loss for epoch 0 0.5755185250059834 The running loss is: 40.93923968449235 The number of items in train is: 28 The loss for epoch 1 1.462115703017584
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-history.jsonl
The running loss is: 17.72650630073622 The number of items in train is: 28 The loss for epoch 2 0.6330895107405793 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-history.jsonl
The running loss is: 13.82603452168405 The number of items in train is: 28 The loss for epoch 3 0.4937869472030018 The running loss is: 12.235369089990854 The number of items in train is: 28 The loss for epoch 4 0.43697746749967337
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-history.jsonl
The running loss is: 12.42129836557433 The number of items in train is: 28 The loss for epoch 5 0.4436177987705118 1
INFO:wandb.wandb_agent:Running runs: ['tc6k32xx']
The running loss is: 11.593332045944408 The number of items in train is: 28 The loss for epoch 6 0.4140475730694431
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-history.jsonl
The running loss is: 12.022986069088802 The number of items in train is: 28 The loss for epoch 7 0.42939235961031436 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-history.jsonl
The running loss is: 11.050711158488411 The number of items in train is: 28 The loss for epoch 8 0.3946682556603004 2 The running loss is: 10.95676115965398 The number of items in train is: 28 The loss for epoch 9 0.3913128985590707 3 Stopping model now Data saved to: 28_May_202005_09PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-history.jsonl
Data saved to: 28_May_202005_09PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 837.403687
66 2020-04-22 sub_region ... 66 830.898865
67 2020-04-23 sub_region ... 67 824.932678
68 2020-04-24 sub_region ... 68 836.498291
69 2020-04-25 sub_region ... 69 841.817627
70 2020-04-26 sub_region ... 70 846.289368
71 2020-04-27 sub_region ... 71 830.351929
72 2020-04-28 sub_region ... 72 852.223999
73 2020-04-29 sub_region ... 73 823.185059
74 2020-04-30 sub_region ... 74 832.774414
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/media/plotly/test_plot_20_754d0a08.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/media/plotly
wandb: Agent Finished Run: tc6k32xx
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_170949-tc6k32xx/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: tc6k32xx INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 7bw2bkzf with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 1 use_mask: False wandb: Agent Started Run: 7bw2bkzf
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/7bw2bkzf INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjdidzJia3pmOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/media/graph/graph_0_summary_0bbaab27.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/media/graph
The running loss is: 15.39689282909967 The number of items in train is: 28 The loss for epoch 0 0.5498890296107025
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-history.jsonl
The running loss is: 44.767726235091686 The number of items in train is: 28 The loss for epoch 1 1.5988473655389888 The running loss is: 22.245736518874764 The number of items in train is: 28 The loss for epoch 2 0.794490589959813 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-history.jsonl
The running loss is: 15.605747589841485 The number of items in train is: 28 The loss for epoch 3 0.5573481282086244 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-history.jsonl
The running loss is: 18.130321642383933 The number of items in train is: 28 The loss for epoch 4 0.6475114872279976 3 Stopping model now Data saved to: 28_May_202005_10PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_10PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['7bw2bkzf']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 410.513519
66 2020-04-22 sub_region ... 66 402.391907
67 2020-04-23 sub_region ... 67 409.347412
68 2020-04-24 sub_region ... 68 416.087708
69 2020-04-25 sub_region ... 69 416.266327
70 2020-04-26 sub_region ... 70 415.997223
71 2020-04-27 sub_region ... 71 405.508575
72 2020-04-28 sub_region ... 72 415.981689
73 2020-04-29 sub_region ... 73 393.018555
74 2020-04-30 sub_region ... 74 405.671448
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-history.jsonl
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/media/plotly/test_plot_10_d4fb56b9.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/media/plotly
wandb: Agent Finished Run: 7bw2bkzf
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/media/plotly/test_plot_all_11_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171004-7bw2bkzf/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 7bw2bkzf INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: tl51mwpo with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 2 use_mask: True wandb: Agent Started Run: tl51mwpo
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/tl51mwpo INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnRsNTFtd3BvOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/media/graph/graph_0_summary_6324096e.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/media/graph
The running loss is: 10.718430198729038 The number of items in train is: 27 The loss for epoch 0 0.3969788962492236 The running loss is: 28.698574017733335 The number of items in train is: 27 The loss for epoch 1 1.0629101488049384
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-history.jsonl
The running loss is: 16.018407717347145 The number of items in train is: 27 The loss for epoch 2 0.5932743599017462 1 The running loss is: 10.534781254827976 The number of items in train is: 27 The loss for epoch 3 0.39017708351214725
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-history.jsonl
2 The running loss is: 5.385364955291152 The number of items in train is: 27 The loss for epoch 4 0.1994579613070797
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-history.jsonl
The running loss is: 7.625352939590812 The number of items in train is: 27 The loss for epoch 5 0.28242047924410413 1 The running loss is: 6.7238668041536584 The number of items in train is: 27 The loss for epoch 6 0.2490321038575429
INFO:wandb.wandb_agent:Running runs: ['tl51mwpo'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-history.jsonl
The running loss is: 7.8772421116009355 The number of items in train is: 27 The loss for epoch 7 0.29174970783707166 The running loss is: 7.317102946341038 The number of items in train is: 27 The loss for epoch 8 0.27100381282744584
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-history.jsonl
1 The running loss is: 5.420254968106747 The number of items in train is: 27 The loss for epoch 9 0.20075018400395359 Data saved to: 28_May_202005_10PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-history.jsonl
Data saved to: 28_May_202005_10PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 779.043823
66 2020-04-22 sub_region ... 66 778.755005
67 2020-04-23 sub_region ... 67 778.881592
68 2020-04-24 sub_region ... 68 779.025024
69 2020-04-25 sub_region ... 69 779.112366
70 2020-04-26 sub_region ... 70 779.083008
71 2020-04-27 sub_region ... 71 778.987671
72 2020-04-28 sub_region ... 72 778.948608
73 2020-04-29 sub_region ... 73 778.723877
74 2020-04-30 sub_region ... 74 778.891602
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: tl51mwpo
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/media/plotly/test_plot_20_03adc1c1.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171015-tl51mwpo/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: tl51mwpo INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: c16hxf83 with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 2 use_mask: False wandb: Agent Started Run: c16hxf83
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/c16hxf83 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmMxNmh4ZjgzOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/media/graph/graph_0_summary_4c786c3f.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/media
The running loss is: 10.668751269578934 The number of items in train is: 27 The loss for epoch 0 0.3951389359103309 The running loss is: 28.521158926188946 The number of items in train is: 27 The loss for epoch 1 1.0563392194884795
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-history.jsonl
The running loss is: 14.176970317959785 The number of items in train is: 27 The loss for epoch 2 0.5250729747392513 The running loss is: 6.591888493858278 The number of items in train is: 27 The loss for epoch 3 0.24414401829104732
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-history.jsonl
1 The running loss is: 7.487370473332703 The number of items in train is: 27 The loss for epoch 4 0.27731001753084084
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-history.jsonl
The running loss is: 5.26125293970108 The number of items in train is: 27 The loss for epoch 5 0.1948612199889289 The running loss is: 6.058157247491181 The number of items in train is: 27 The loss for epoch 6 0.22437619435152523
INFO:wandb.wandb_agent:Running runs: ['c16hxf83']
1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-history.jsonl
The running loss is: 4.886977478861809 The number of items in train is: 27 The loss for epoch 7 0.18099916588377069 The running loss is: 8.049873136915267 The number of items in train is: 27 The loss for epoch 8 0.29814344951538024
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-history.jsonl
1 The running loss is: 6.706313902512193 The number of items in train is: 27 The loss for epoch 9 0.24838199638934047 2 Data saved to: 28_May_202005_10PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-history.jsonl
Data saved to: 28_May_202005_10PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 755.005249
66 2020-04-22 sub_region ... 66 751.443115
67 2020-04-23 sub_region ... 67 753.844849
68 2020-04-24 sub_region ... 68 755.624390
69 2020-04-25 sub_region ... 69 755.841370
70 2020-04-26 sub_region ... 70 755.883606
71 2020-04-27 sub_region ... 71 752.995850
72 2020-04-28 sub_region ... 72 755.255066
73 2020-04-29 sub_region ... 73 751.328247
74 2020-04-30 sub_region ... 74 753.881714
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/media/plotly/test_plot_20_9c0dfd36.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: c16hxf83
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171030-c16hxf83/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: c16hxf83 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: vllrz9kx with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 3 use_mask: True wandb: Agent Started Run: vllrz9kx
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/vllrz9kx INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnZsbHJ6OWt4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/media/graph/graph_0_summary_ea5c822b.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/media
The running loss is: 23.849574618041515 The number of items in train is: 27 The loss for epoch 0 0.883317578445982 The running loss is: 20.568836465477943 The number of items in train is: 27 The loss for epoch 1 0.7618087579806646
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-history.jsonl
The running loss is: 19.27038937062025 The number of items in train is: 27 The loss for epoch 2 0.7137181248377871
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-history.jsonl
The running loss is: 17.25364562869072 The number of items in train is: 27 The loss for epoch 3 0.6390239121737303 1 The running loss is: 14.78194186091423 The number of items in train is: 27 The loss for epoch 4 0.5474793281820085
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-history.jsonl
The running loss is: 10.99204146116972 The number of items in train is: 27 The loss for epoch 5 0.40711264670998965 1
INFO:wandb.wandb_agent:Running runs: ['vllrz9kx']
The running loss is: 34.4778664149344 The number of items in train is: 27 The loss for epoch 6 1.2769580153679405
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-history.jsonl
The running loss is: 18.340780273079872 The number of items in train is: 27 The loss for epoch 7 0.6792881582622174 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-history.jsonl
The running loss is: 18.76011623442173 The number of items in train is: 27 The loss for epoch 8 0.6948191197933974 2 The running loss is: 18.743604812771082 The number of items in train is: 27 The loss for epoch 9 0.6942075856581882 Data saved to: 28_May_202005_10PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-history.jsonl
Data saved to: 28_May_202005_10PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 351.287323
66 2020-04-22 sub_region ... 66 351.352722
67 2020-04-23 sub_region ... 67 351.261353
68 2020-04-24 sub_region ... 68 351.227020
69 2020-04-25 sub_region ... 69 351.256317
70 2020-04-26 sub_region ... 70 351.258667
71 2020-04-27 sub_region ... 71 351.388000
72 2020-04-28 sub_region ... 72 351.195374
73 2020-04-29 sub_region ... 73 351.413818
74 2020-04-30 sub_region ... 74 351.286987
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/media/plotly/test_plot_20_3234c249.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/media/plotly
wandb: Agent Finished Run: vllrz9kx
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171046-vllrz9kx/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: vllrz9kx INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 8hosk45y with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 3 use_mask: False wandb: Agent Started Run: 8hosk45y
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/8hosk45y INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjhob3NrNDV5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/media/graph/graph_0_summary_08442c6e.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/media
The running loss is: 23.785254307091236 The number of items in train is: 27 The loss for epoch 0 0.8809353447070828 The running loss is: 20.365187123417854 The number of items in train is: 27 The loss for epoch 1 0.7542661897562168
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-history.jsonl
The running loss is: 19.407820589840412 The number of items in train is: 27 The loss for epoch 2 0.7188081699940894
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-history.jsonl
The running loss is: 16.047366067767143 The number of items in train is: 27 The loss for epoch 3 0.594346891398783 1 The running loss is: 16.049381844699383 The number of items in train is: 27 The loss for epoch 4 0.5944215498036809
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-history.jsonl
The running loss is: 8.695411691442132 The number of items in train is: 27 The loss for epoch 5 0.3220522848682271 1
INFO:wandb.wandb_agent:Running runs: ['8hosk45y']
The running loss is: 11.989098764955997 The number of items in train is: 27 The loss for epoch 6 0.4440406949983703
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-history.jsonl
The running loss is: 12.975145380944014 The number of items in train is: 27 The loss for epoch 7 0.48056094003496347 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-history.jsonl
The running loss is: 6.981494817882776 The number of items in train is: 27 The loss for epoch 8 0.25857388214380655 The running loss is: 7.491051498800516 The number of items in train is: 27 The loss for epoch 9 0.27744635180742655 1 Data saved to: 28_May_202005_11PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_11PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 709.696960
66 2020-04-22 sub_region ... 66 709.161499
67 2020-04-23 sub_region ... 67 708.643188
68 2020-04-24 sub_region ... 68 708.597656
69 2020-04-25 sub_region ... 69 708.908020
70 2020-04-26 sub_region ... 70 708.755493
71 2020-04-27 sub_region ... 71 709.836182
72 2020-04-28 sub_region ... 72 708.476990
73 2020-04-29 sub_region ... 73 709.191528
74 2020-04-30 sub_region ... 74 708.709473
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/media/plotly/test_plot_20_509b02c0.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 8hosk45y
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171103-8hosk45y/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 8hosk45y INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: x293l7uz with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 4 use_mask: True wandb: Agent Started Run: x293l7uz
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/x293l7uz INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOngyOTNsN3V6OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/media/graph/graph_0_summary_4b05e63a.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/media/graph
The running loss is: 25.31017677485943 The number of items in train is: 26 The loss for epoch 0 0.9734683374945934 The running loss is: 17.691114902496338 The number of items in train is: 26 The loss for epoch 1 0.6804274962498591 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-history.jsonl
The running loss is: 21.284134477376938 The number of items in train is: 26 The loss for epoch 2 0.8186205568221899 The running loss is: 14.811767756938934 The number of items in train is: 26 The loss for epoch 3 0.5696833752668821
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-history.jsonl
The running loss is: 12.015214286744595 The number of items in train is: 26 The loss for epoch 4 0.46212362641325366 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-history.jsonl
The running loss is: 10.353260762989521 The number of items in train is: 26 The loss for epoch 5 0.3982023370380585 2 The running loss is: 7.667264841496944 The number of items in train is: 26 The loss for epoch 6 0.2948948015960363 3 Stopping model now Data saved to: 28_May_202005_11PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['x293l7uz'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-history.jsonl
Data saved to: 28_May_202005_11PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 318.850616
66 2020-04-22 sub_region ... 66 316.706177
67 2020-04-23 sub_region ... 67 320.013977
68 2020-04-24 sub_region ... 68 321.565765
69 2020-04-25 sub_region ... 69 321.352051
70 2020-04-26 sub_region ... 70 320.382446
71 2020-04-27 sub_region ... 71 318.602081
72 2020-04-28 sub_region ... 72 318.317017
73 2020-04-29 sub_region ... 73 318.346680
74 2020-04-30 sub_region ... 74 320.484741
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: x293l7uz
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/media/plotly/test_plot_all_15_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/media/plotly/test_plot_14_b1c54b64.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171119-x293l7uz/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: x293l7uz INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: tozkkomi with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 4 use_mask: False wandb: Agent Started Run: tozkkomi
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/tozkkomi INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnRvemtrb21pOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/config.yaml
The running loss is: 25.090075597167015 The number of items in train is: 26
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-summary.json
The loss for epoch 0
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-events.jsonl
0.9650029075833467
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/media/graph/graph_0_summary_38ea3612.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/media
The running loss is: 17.45224541425705 The number of items in train is: 26 The loss for epoch 1 0.6712402082406558 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-history.jsonl
The running loss is: 21.34785506129265 The number of items in train is: 26 The loss for epoch 2 0.8210713485112557 The running loss is: 14.617094554007053 The number of items in train is: 26 The loss for epoch 3 0.5621959443848866 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-history.jsonl
The running loss is: 12.857044279575348 The number of items in train is: 26 The loss for epoch 4 0.4945017030605903
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-history.jsonl
The running loss is: 10.908861950039864 The number of items in train is: 26 The loss for epoch 5 0.4195716134630717 1 The running loss is: 7.671748608350754 The number of items in train is: 26 The loss for epoch 6 0.2950672541673367
INFO:wandb.wandb_agent:Running runs: ['tozkkomi'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-history.jsonl
The running loss is: 6.437005430459976 The number of items in train is: 26 The loss for epoch 7 0.2475771319407683 The running loss is: 7.865247845649719 The number of items in train is: 26 The loss for epoch 8 0.3025095325249892 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-history.jsonl
The running loss is: 5.2042678743600845 The number of items in train is: 26 The loss for epoch 9 0.2001641490138494 Data saved to: 28_May_202005_11PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_11PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 570.512756
66 2020-04-22 sub_region ... 66 571.143372
67 2020-04-23 sub_region ... 67 570.903015
68 2020-04-24 sub_region ... 68 570.691162
69 2020-04-25 sub_region ... 69 570.595093
70 2020-04-26 sub_region ... 70 570.591003
71 2020-04-27 sub_region ... 71 570.901611
72 2020-04-28 sub_region ... 72 570.663696
73 2020-04-29 sub_region ... 73 571.259827
74 2020-04-30 sub_region ... 74 571.026001
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/media/plotly/test_plot_20_96e49c04.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: tozkkomi
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171134-tozkkomi/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: tozkkomi INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: b7f459dg with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 5 use_mask: True wandb: Agent Started Run: b7f459dg
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/b7f459dg INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmI3ZjQ1OWRnOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/media/graph/graph_0_summary_939c93f5.graph.json
The running loss is: 24.998789221048355 The number of items in train is:
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/requirements.txt
26
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-metadata.json
The loss for epoch 0
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-events.jsonl
0.9614918931172445
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/media/graph
The running loss is: 19.62556444108486 The number of items in train is: 26 The loss for epoch 1 0.754829401580187 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-history.jsonl
The running loss is: 19.021592140197754 The number of items in train is: 26 The loss for epoch 2 0.7315996976999136 The running loss is: 13.53062754869461 The number of items in train is: 26 The loss for epoch 3 0.5204087518728696
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-history.jsonl
The running loss is: 11.939093425869942 The number of items in train is: 26 The loss for epoch 4 0.45919590099499774
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-history.jsonl
The running loss is: 12.370361641049385 The number of items in train is: 26 The loss for epoch 5 0.475783140040361 1 The running loss is: 10.105389013886452 The number of items in train is: 26 The loss for epoch 6 0.388668808226402
INFO:wandb.wandb_agent:Running runs: ['b7f459dg'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-history.jsonl
The running loss is: 5.374523870646954 The number of items in train is: 26 The loss for epoch 7 0.20671245656334436 1 The running loss is: 4.875691753812134 The number of items in train is: 26 The loss for epoch 8 0.18752660591585132
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-history.jsonl
The running loss is: 7.123469600221142 The number of items in train is: 26 The loss for epoch 9 0.27397960000850546 1 Data saved to: 28_May_202005_11PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_11PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 697.734924
66 2020-04-22 sub_region ... 66 699.246887
67 2020-04-23 sub_region ... 67 698.447754
68 2020-04-24 sub_region ... 68 697.592102
69 2020-04-25 sub_region ... 69 697.349731
70 2020-04-26 sub_region ... 70 697.215332
71 2020-04-27 sub_region ... 71 698.537048
72 2020-04-28 sub_region ... 72 697.177002
73 2020-04-29 sub_region ... 73 699.459961
74 2020-04-30 sub_region ... 74 698.878723
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: b7f459dg
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/media/plotly/test_plot_20_70b23860.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171151-b7f459dg/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: b7f459dg INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: de4cq8v7 with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 2 out_seq_length: 5 use_mask: False wandb: Agent Started Run: de4cq8v7
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/de4cq8v7 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmRlNGNxOHY3OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/media/graph/graph_0_summary_e281b7e9.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/media/graph
The running loss is: 24.827137678861618 The number of items in train is: 26 The loss for epoch 0 0.9548899107254468 The running loss is: 19.719864428043365 The number of items in train is: 26 The loss for epoch 1 0.758456324155514 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-history.jsonl
The running loss is: 18.63474379479885 The number of items in train is: 26 The loss for epoch 2 0.7167209151845712 The running loss is: 13.806637480854988 The number of items in train is: 26 The loss for epoch 3 0.5310245184944227 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-history.jsonl
The running loss is: 11.331828974187374 The number of items in train is: 26 The loss for epoch 4 0.4358395759302836
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-history.jsonl
The running loss is: 10.775343157351017 The number of items in train is: 26 The loss for epoch 5 0.4144362752827314 1 The running loss is: 10.32153557986021 The number of items in train is: 26 The loss for epoch 6 0.3969821376869312 2
INFO:wandb.wandb_agent:Running runs: ['de4cq8v7'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-history.jsonl
The running loss is: 7.037654765415937 The number of items in train is: 26 The loss for epoch 7 0.2706790294390745 The running loss is: 14.718151992186904 The number of items in train is: 26 The loss for epoch 8 0.5660827689302655 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-history.jsonl
The running loss is: 5.737017912324518 The number of items in train is: 26 The loss for epoch 9 0.22065453508940452 Data saved to: 28_May_202005_12PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_12PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 632.474121
66 2020-04-22 sub_region ... 66 634.329102
67 2020-04-23 sub_region ... 67 632.773071
68 2020-04-24 sub_region ... 68 631.713074
69 2020-04-25 sub_region ... 69 631.196228
70 2020-04-26 sub_region ... 70 631.151917
71 2020-04-27 sub_region ... 71 632.160889
72 2020-04-28 sub_region ... 72 631.541016
73 2020-04-29 sub_region ... 73 633.977478
74 2020-04-30 sub_region ... 74 632.605713
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/media/plotly/test_plot_20_a4b0535b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: de4cq8v7
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171208-de4cq8v7/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: de4cq8v7 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: rm3rj91b with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 1 use_mask: True wandb: Agent Started Run: rm3rj91b
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/rm3rj91b INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnJtM3JqOTFiOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/media/graph/graph_0_summary_afe148b1.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/media/graph
The running loss is: 16.056726850569248 The number of items in train is: 28 The loss for epoch 0 0.5734545303774732
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-summary.json
The running loss is: 46.208357490599155 The number of items in train is: 28 The loss for epoch 1 1.6502984818071127
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-summary.json
The running loss is: 29.032308392226696 The number of items in train is: 28 The loss for epoch 2 1.0368681568652391
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-summary.json
The running loss is: 28.841471418738365 The number of items in train is: 28 The loss for epoch 3 1.0300525506692273
INFO:wandb.wandb_agent:Running runs: ['rm3rj91b'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-summary.json
The running loss is: 28.338371301069856 The number of items in train is: 28 The loss for epoch 4 1.0120846893239235 1 The running loss is: 28.69572694809176 The number of items in train is: 28 The loss for epoch 5 1.0248473910032772 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-summary.json
The running loss is: 19.646056853234768 The number of items in train is: 28 The loss for epoch 6 0.7016448876155275 3 Stopping model now Data saved to: 28_May_202005_12PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-summary.json
Data saved to: 28_May_202005_12PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 429.154022
66 2020-04-22 sub_region ... 66 428.729492
67 2020-04-23 sub_region ... 67 429.150635
68 2020-04-24 sub_region ... 68 429.542450
69 2020-04-25 sub_region ... 69 429.516174
70 2020-04-26 sub_region ... 70 429.553772
71 2020-04-27 sub_region ... 71 428.790649
72 2020-04-28 sub_region ... 72 429.774963
73 2020-04-29 sub_region ... 73 428.176636
74 2020-04-30 sub_region ... 74 428.942596
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/media/plotly/test_plot_14_bdfadf79.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/media/plotly
wandb: Agent Finished Run: rm3rj91b
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/media/plotly/test_plot_all_15_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171226-rm3rj91b/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: rm3rj91b INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 6pb0wrjo with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 1 use_mask: False wandb: Agent Started Run: 6pb0wrjo
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/6pb0wrjo INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjZwYjB3cmpvOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/media/graph/graph_0_summary_5cc08279.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/media
The running loss is: 13.879746047779918 The number of items in train is: 28 The loss for epoch 0 0.4957052159921399
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-summary.json
The running loss is: 50.16468261182308 The number of items in train is: 28 The loss for epoch 1 1.7915958075651102
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-summary.json
The running loss is: 28.752860818058252 The number of items in train is: 28 The loss for epoch 2 1.0268878863592232
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-summary.json
The running loss is: 28.75844579562545 The number of items in train is: 28 The loss for epoch 3 1.027087349843766
INFO:wandb.wandb_agent:Running runs: ['6pb0wrjo'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-summary.json
The running loss is: 28.228818399831653 The number of items in train is: 28 The loss for epoch 4 1.0081720857082732 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-summary.json
The running loss is: 28.275303000351414 The number of items in train is: 28 The loss for epoch 5 1.0098322500125505 2 The running loss is: 19.328091056086123 The number of items in train is: 28 The loss for epoch 6 0.6902889662887901 3 Stopping model now
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-summary.json
Data saved to: 28_May_202005_12PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_12PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/config.yaml /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 429.017181
66 2020-04-22 sub_region ... 66 428.986389
67 2020-04-23 sub_region ... 67 429.089111
68 2020-04-24 sub_region ... 68 429.224731
69 2020-04-25 sub_region ... 69 429.164246
70 2020-04-26 sub_region ... 70 429.191895
71 2020-04-27 sub_region ... 71 428.905151
72 2020-04-28 sub_region ... 72 429.181915
73 2020-04-29 sub_region ... 73 428.645477
74 2020-04-30 sub_region ... 74 428.955353
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/media/plotly/test_plot_14_562144b7.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 6pb0wrjo
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/media/plotly/test_plot_all_15_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171244-6pb0wrjo/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 6pb0wrjo INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: o25lms00 with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 2 use_mask: True wandb: Agent Started Run: o25lms00
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/o25lms00 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-metadata.json INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm8yNWxtczAwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/media/graph/graph_0_summary_feea31d7.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/media/graph
The running loss is: 9.515796359628439 The number of items in train is: 27 The loss for epoch 0 0.3524369022084607
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-history.jsonl
The running loss is: 31.958917625248432 The number of items in train is: 27 The loss for epoch 1 1.1836636157499418
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-history.jsonl
The running loss is: 17.400996170938015 The number of items in train is: 27 The loss for epoch 2 0.6444813396643709 1
INFO:wandb.wandb_agent:Running runs: ['o25lms00'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-history.jsonl
The running loss is: 17.490230657160282 The number of items in train is: 27 The loss for epoch 3 0.647786320635566 The running loss is: 17.88597560673952 The number of items in train is: 27 The loss for epoch 4 0.6624435409903526
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-history.jsonl
The running loss is: 17.275350011885166 The number of items in train is: 27 The loss for epoch 5 0.6398277782179691
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-history.jsonl
The running loss is: 17.80729231238365 The number of items in train is: 27 The loss for epoch 6 0.6595293449030982
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-history.jsonl
The running loss is: 17.339848961681128 The number of items in train is: 27 The loss for epoch 7 0.6422166282104121
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-history.jsonl
The running loss is: 16.749763071537018 The number of items in train is: 27 The loss for epoch 8 0.6203615952421118
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-history.jsonl
The running loss is: 17.160026656463742 The number of items in train is: 27 The loss for epoch 9 0.6355565428319905 Data saved to: 28_May_202005_13PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_13PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-history.jsonl
CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 452.964355
66 2020-04-22 sub_region ... 66 448.373505
67 2020-04-23 sub_region ... 67 451.731903
68 2020-04-24 sub_region ... 68 454.205078
69 2020-04-25 sub_region ... 69 454.852173
70 2020-04-26 sub_region ... 70 454.945068
71 2020-04-27 sub_region ... 71 451.557831
72 2020-04-28 sub_region ... 72 453.470245
73 2020-04-29 sub_region ... 73 448.514465
74 2020-04-30 sub_region ... 74 451.503876
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/media/plotly/test_plot_20_da2b038c.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: o25lms00
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171259-o25lms00/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: o25lms00 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: de1bsktw with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 2 use_mask: False wandb: Agent Started Run: de1bsktw
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/de1bsktw INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmRlMWJza3R3OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/media/graph/graph_0_summary_f9cc4e35.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/media/graph
The running loss is: 9.540501032024622 The number of items in train is: 27 The loss for epoch 0 0.353351890074986
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-summary.json
The running loss is: 32.01947431452572 The number of items in train is: 27 The loss for epoch 1 1.1859064560935453
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-summary.json
The running loss is: 17.433912321925163 The number of items in train is: 27 The loss for epoch 2 0.6457004563675987 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-summary.json
The running loss is: 17.372653737664223 The number of items in train is: 27 The loss for epoch 3 0.6434316199134897
INFO:wandb.wandb_agent:Running runs: ['de1bsktw']
The running loss is: 17.937954049557447 The number of items in train is: 27 The loss for epoch 4 0.6643686685021277
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-summary.json
The running loss is: 17.54516338557005 The number of items in train is: 27 The loss for epoch 5 0.649820866132224
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-summary.json
The running loss is: 17.718922704458237 The number of items in train is: 27 The loss for epoch 6 0.6562563964614162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-summary.json
The running loss is: 17.30660080164671 The number of items in train is: 27 The loss for epoch 7 0.6409852148758041
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-summary.json
The running loss is: 16.718926943838596 The number of items in train is: 27 The loss for epoch 8 0.6192195164384665
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-summary.json
The running loss is: 16.827449632808566 The number of items in train is: 27 The loss for epoch 9 0.6232388752892062 Data saved to: 28_May_202005_13PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_13PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 457.992615
66 2020-04-22 sub_region ... 66 454.040710
67 2020-04-23 sub_region ... 67 456.812286
68 2020-04-24 sub_region ... 68 459.140167
69 2020-04-25 sub_region ... 69 459.434723
70 2020-04-26 sub_region ... 70 459.551544
71 2020-04-27 sub_region ... 71 456.045074
72 2020-04-28 sub_region ... 72 458.655548
73 2020-04-29 sub_region ... 73 453.452576
74 2020-04-30 sub_region ... 74 456.783661
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: de1bsktw
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/media/plotly/test_plot_20_c566c54f.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171320-de1bsktw/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: de1bsktw INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: hb7wkf8z with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 3 use_mask: True wandb: Agent Started Run: hb7wkf8z
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/hb7wkf8z INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmhiN3drZjh6OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/media/graph/graph_0_summary_0b1d345c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/media
The running loss is: 31.62208504229784 The number of items in train is: 27 The loss for epoch 0 1.17118833489992
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-history.jsonl
The running loss is: 19.408665716648102 The number of items in train is: 27 The loss for epoch 1 0.7188394709869668 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-history.jsonl
The running loss is: 20.407186299562454 The number of items in train is: 27 The loss for epoch 2 0.7558217147986094
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-history.jsonl
The running loss is: 20.506580412387848 The number of items in train is: 27 The loss for epoch 3 0.759502978236587
INFO:wandb.wandb_agent:Running runs: ['hb7wkf8z']
The running loss is: 20.30820331722498 The number of items in train is: 27 The loss for epoch 4 0.7521556784157399
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-history.jsonl
The running loss is: 19.309663198888302 The number of items in train is: 27 The loss for epoch 5 0.7151727110699371 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-history.jsonl
The running loss is: 18.341759957373142 The number of items in train is: 27 The loss for epoch 6 0.679324442865672
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-history.jsonl
The running loss is: 17.737579837441444 The number of items in train is: 27 The loss for epoch 7 0.6569474013867201 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-history.jsonl
The running loss is: 19.28192499279976 The number of items in train is: 27 The loss for epoch 8 0.7141453701036947
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-history.jsonl
The running loss is: 18.758581429719925 The number of items in train is: 27 The loss for epoch 9 0.694762275174812 1 Data saved to: 28_May_202005_13PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_13PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 337.341278
66 2020-04-22 sub_region ... 66 337.345581
67 2020-04-23 sub_region ... 67 337.355377
68 2020-04-24 sub_region ... 68 337.360901
69 2020-04-25 sub_region ... 69 337.360565
70 2020-04-26 sub_region ... 70 337.355988
71 2020-04-27 sub_region ... 71 337.354065
72 2020-04-28 sub_region ... 72 337.343140
73 2020-04-29 sub_region ... 73 337.349548
74 2020-04-30 sub_region ... 74 337.355652
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/media/plotly/test_plot_20_e91edfe1.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: hb7wkf8z
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171341-hb7wkf8z/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: hb7wkf8z INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: giily2eq with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 3 use_mask: False wandb: Agent Started Run: giily2eq
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/giily2eq INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmdpaWx5MmVxOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/media/graph/graph_0_summary_ae0c622c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/media
The running loss is: 31.686953715980053 The number of items in train is: 27 The loss for epoch 0 1.1735908783696316
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-summary.json
The running loss is: 19.326208144426346 The number of items in train is: 27 The loss for epoch 1 0.7157854868306054 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-summary.json
The running loss is: 20.44809077680111 The number of items in train is: 27 The loss for epoch 2 0.7573366954370782
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-summary.json
The running loss is: 20.70396327972412 The number of items in train is: 27 The loss for epoch 3 0.7668134548045971
INFO:wandb.wandb_agent:Running runs: ['giily2eq'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-summary.json
The running loss is: 20.114036343991756 The number of items in train is: 27 The loss for epoch 4 0.7449643090367317 1 The running loss is: 19.89861784130335 The number of items in train is: 27 The loss for epoch 5 0.7369858459741981
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-summary.json
The running loss is: 18.967432893812656 The number of items in train is: 27 The loss for epoch 6 0.7024975145856539 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-summary.json
The running loss is: 19.17745415121317 The number of items in train is: 27 The loss for epoch 7 0.7102760796745619
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-summary.json
The running loss is: 19.25604397803545 The number of items in train is: 27 The loss for epoch 8 0.713186814001313
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-summary.json
The running loss is: 18.80481818318367 The number of items in train is: 27 The loss for epoch 9 0.6964747475253211 Data saved to: 28_May_202005_14PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_14PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 353.958618
66 2020-04-22 sub_region ... 66 353.994934
67 2020-04-23 sub_region ... 67 353.966064
68 2020-04-24 sub_region ... 68 353.948608
69 2020-04-25 sub_region ... 69 353.950348
70 2020-04-26 sub_region ... 70 353.951172
71 2020-04-27 sub_region ... 71 353.985535
72 2020-04-28 sub_region ... 72 353.945618
73 2020-04-29 sub_region ... 73 354.007202
74 2020-04-30 sub_region ... 74 353.972107
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/media/plotly/test_plot_20_384740d4.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: giily2eq
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171401-giily2eq/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: giily2eq INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 27ymc3gp with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 4 use_mask: True wandb: Agent Started Run: 27ymc3gp
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/27ymc3gp INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjI3eW1jM2dwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/media/graph/graph_0_summary_32804e1d.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/media
The running loss is: 28.677166178822517 The number of items in train is: 26 The loss for epoch 0 1.1029679299547122
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-history.jsonl
The running loss is: 19.185990750789642 The number of items in train is: 26 The loss for epoch 1 0.737922721184217
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-history.jsonl
The running loss is: 20.213715583086014 The number of items in train is: 26 The loss for epoch 2 0.777450599349462 The running loss is: 15.579263865947723 The number of items in train is: 26 The loss for epoch 3 0.5992024563826047
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-history.jsonl
1
INFO:wandb.wandb_agent:Running runs: ['27ymc3gp']
The running loss is: 15.614615187048912 The number of items in train is: 26 The loss for epoch 4 0.6005621225788043
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-history.jsonl
The running loss is: 14.192182525992393 The number of items in train is: 26 The loss for epoch 5 0.5458531740766305 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-history.jsonl
The running loss is: 17.893233135342598 The number of items in train is: 26 The loss for epoch 6 0.6882012744362538
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-history.jsonl
The running loss is: 15.887526050209999 The number of items in train is: 26 The loss for epoch 7 0.6110586942388461
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-history.jsonl
The running loss is: 13.056910991668701 The number of items in train is: 26 The loss for epoch 8 0.50218888429495 1 The running loss is: 10.211280450224876 The number of items in train is: 26 The loss for epoch 9 0.39274155577787984 2 Data saved to: 28_May_202005_14PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-history.jsonl
Data saved to: 28_May_202005_14PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 409.355835
66 2020-04-22 sub_region ... 66 409.509674
67 2020-04-23 sub_region ... 67 409.471558
68 2020-04-24 sub_region ... 68 409.399261
69 2020-04-25 sub_region ... 69 409.354004
70 2020-04-26 sub_region ... 70 409.365173
71 2020-04-27 sub_region ... 71 409.424011
72 2020-04-28 sub_region ... 72 409.376404
73 2020-04-29 sub_region ... 73 409.552612
74 2020-04-30 sub_region ... 74 409.473022
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/media/plotly/test_plot_20_eefb9a0f.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 27ymc3gp
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171420-27ymc3gp/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 27ymc3gp INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 5fwqi39m with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 4 use_mask: False wandb: Agent Started Run: 5fwqi39m
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/5fwqi39m INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjVmd3FpMzltOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/media/graph/graph_0_summary_89a8d80e.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/media
The running loss is: 28.61015349626541 The number of items in train is: 26 The loss for epoch 0 1.1003905190871313
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-summary.json
The running loss is: 19.129196166992188 The number of items in train is: 26 The loss for epoch 1 0.7357383141150842
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-summary.json
The running loss is: 20.223093792796135 The number of items in train is: 26 The loss for epoch 2 0.7778112997229283 The running loss is: 15.592137679457664 The number of items in train is: 26
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-history.jsonl
The loss for epoch 3
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-summary.json
0.599697603056064 1
INFO:wandb.wandb_agent:Running runs: ['5fwqi39m']
The running loss is: 14.741975575685501 The number of items in train is: 26 The loss for epoch 4 0.5669990606032885
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-summary.json
The running loss is: 14.141681730747223 The number of items in train is: 26 The loss for epoch 5 0.5439108357979701 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-summary.json
The running loss is: 17.65192748606205 The number of items in train is: 26 The loss for epoch 6 0.6789202879254634
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-summary.json
The running loss is: 13.934842199087143 The number of items in train is: 26 The loss for epoch 7 0.5359554691956594 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-summary.json
The running loss is: 10.86246071755886 The number of items in train is: 26 The loss for epoch 8 0.4177869506753408 The running loss is: 9.102634258568287 The number of items in train is: 26 The loss for epoch 9 0.3501013176372418 1 Data saved to: 28_May_202005_14PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-summary.json
Data saved to: 28_May_202005_14PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 475.105896
66 2020-04-22 sub_region ... 66 475.061249
67 2020-04-23 sub_region ... 67 475.085632
68 2020-04-24 sub_region ... 68 475.112213
69 2020-04-25 sub_region ... 69 475.114807
70 2020-04-26 sub_region ... 70 475.139618
71 2020-04-27 sub_region ... 71 475.079346
72 2020-04-28 sub_region ... 72 475.173126
73 2020-04-29 sub_region ... 73 475.043427
74 2020-04-30 sub_region ... 74 475.085358
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/media/plotly/test_plot_20_114ca43c.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 5fwqi39m
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171441-5fwqi39m/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 5fwqi39m INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 2d2cmon8 with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 5 use_mask: True wandb: Agent Started Run: 2d2cmon8
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/2d2cmon8 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjJkMmNtb244OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/media/graph/graph_0_summary_8a875c68.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/media
The running loss is: 26.287698671221733 The number of items in train is: 26 The loss for epoch 0 1.0110653335085282
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-history.jsonl
The running loss is: 22.019511476159096 The number of items in train is: 26 The loss for epoch 1 0.8469042875445806
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-history.jsonl
The running loss is: 19.7527959048748 The number of items in train is: 26 The loss for epoch 2 0.7597229194182616 The running loss is: 15.987582370638847 The number of items in train is: 26 The loss for epoch 3 0.6149070142553403
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-history.jsonl
1 The running loss is: 15.31729331612587 The number of items in train is: 26
INFO:wandb.wandb_agent:Running runs: ['2d2cmon8']
The loss for epoch 4 0.5891266660048411
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-history.jsonl
The running loss is: 13.650041669607162 The number of items in train is: 26 The loss for epoch 5 0.5250016026771985 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-history.jsonl
The running loss is: 13.348268389701843 The number of items in train is: 26 The loss for epoch 6 0.5133949380654556 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-history.jsonl
The running loss is: 11.254892647266388 The number of items in train is: 26 The loss for epoch 7 0.4328804864333226 3 Stopping model now Data saved to: 28_May_202005_15PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_15PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-history.jsonl
CSV Path below United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/config.yaml
torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 347.929901
66 2020-04-22 sub_region ... 66 348.376862
67 2020-04-23 sub_region ... 67 348.171692
68 2020-04-24 sub_region ... 68 347.880554
69 2020-04-25 sub_region ... 69 347.761993
70 2020-04-26 sub_region ... 70 347.774658
71 2020-04-27 sub_region ... 71 348.046265
72 2020-04-28 sub_region ... 72 347.813934
73 2020-04-29 sub_region ... 73 348.664154
74 2020-04-30 sub_region ... 74 348.247253
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/media/plotly/test_plot_16_3f181331.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 2d2cmon8
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/media/plotly/test_plot_all_17_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171503-2d2cmon8/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 2d2cmon8 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: qlrtjgmw with config: batch_size: 2 forecast_history: 10 lr: 0.001 number_encoder_layers: 3 out_seq_length: 5 use_mask: False wandb: Agent Started Run: qlrtjgmw
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/qlrtjgmw INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnFscnRqZ213OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/media/graph/graph_0_summary_c8fe355a.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/media/graph
The running loss is: 26.332314282655716 The number of items in train is: 26 The loss for epoch 0 1.0127813185636814
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-history.jsonl
The running loss is: 21.822422578930855 The number of items in train is: 26 The loss for epoch 1 0.8393239453434944
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-history.jsonl
The running loss is: 19.969776809215546 The number of items in train is: 26 The loss for epoch 2 0.7680683388159826 The running loss is: 15.921830900013447 The number of items in train is: 26 The loss for epoch 3 0.6123781115389787
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-history.jsonl
1 The running loss is: 15.494909837841988 The number of items in train is: 26 The loss for epoch 4 0.5959580706862303
INFO:wandb.wandb_agent:Running runs: ['qlrtjgmw'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-history.jsonl
The running loss is: 14.208502128720284 The number of items in train is: 26 The loss for epoch 5 0.5464808511046263 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-history.jsonl
The running loss is: 16.151983082294464 The number of items in train is: 26 The loss for epoch 6 0.621230118549787 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-history.jsonl
The running loss is: 12.41115165501833 The number of items in train is: 26 The loss for epoch 7 0.4773519867314742 The running loss is: 9.447976488620043 The number of items in train is: 26 The loss for epoch 8 0.3633837111007709
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-history.jsonl
The running loss is: 9.989659074693918 The number of items in train is: 26 The loss for epoch 9 0.3842176567189969 1 Data saved to: 28_May_202005_15PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-history.jsonl
Data saved to: 28_May_202005_15PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 578.508301
66 2020-04-22 sub_region ... 66 578.518188
67 2020-04-23 sub_region ... 67 578.516479
68 2020-04-24 sub_region ... 68 578.510986
69 2020-04-25 sub_region ... 69 578.503906
70 2020-04-26 sub_region ... 70 578.503906
71 2020-04-27 sub_region ... 71 578.500427
72 2020-04-28 sub_region ... 72 578.506470
73 2020-04-29 sub_region ... 73 578.516052
74 2020-04-30 sub_region ... 74 578.512817
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/media/plotly/test_plot_20_ae1f0e4f.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/media/plotly
wandb: Agent Finished Run: qlrtjgmw
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171519-qlrtjgmw/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: qlrtjgmw INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: t3gzqxhh with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 1 use_mask: True wandb: Agent Started Run: t3gzqxhh
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/t3gzqxhh INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnQzZ3pxeGhoOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
15.545928395004012
The number of items in train is:
28
The loss for epoch 0
0.5552117283930004
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/media/graph/graph_0_summary_187e9097.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/media/graph
The running loss is: 21.661523299873807 The number of items in train is: 28 The loss for epoch 1 0.7736258321383502 The running loss is: 14.07579830638133 The number of items in train is: 28 The loss for epoch 2 0.5027070823707618 The running loss is: 11.572435699170455
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-history.jsonl
The number of items in train is: 28 The loss for epoch 3 0.4133012749703734 1 The running loss is: 11.783770642941818 The number of items in train is: 28 The loss for epoch 4 0.42084895153363633 The running loss is: 9.61011728970334 The number of items in train is: 28 The loss for epoch 5 0.3432184746322621 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-history.jsonl
The running loss is: 13.171849727863446 The number of items in train is: 28 The loss for epoch 6 0.4704232045665516 The running loss is: 9.562567624612711 The number of items in train is: 28 The loss for epoch 7 0.34152027230759685 1 The running loss is: 11.004484111152124 The number of items in train is: 28 The loss for epoch 8 0.3930172896840044 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-history.jsonl
The running loss is: 10.858391232584836 The number of items in train is: 28 The loss for epoch 9 0.38779968687802985 Data saved to: 28_May_202005_15PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_15PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['t3gzqxhh'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 856.106934
66 2020-04-22 sub_region ... 66 898.349731
67 2020-04-23 sub_region ... 67 857.475891
68 2020-04-24 sub_region ... 68 861.468872
69 2020-04-25 sub_region ... 69 885.428162
70 2020-04-26 sub_region ... 70 897.812500
71 2020-04-27 sub_region ... 71 894.293274
72 2020-04-28 sub_region ... 72 929.067383
73 2020-04-29 sub_region ... 73 878.263550
74 2020-04-30 sub_region ... 74 869.549072
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/media/plotly/test_plot_20_314de4b9.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: t3gzqxhh
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171535-t3gzqxhh/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: t3gzqxhh INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: qludt1wz with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 1 use_mask: False wandb: Agent Started Run: qludt1wz
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/qludt1wz INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnFsdWR0MXd6OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
15.36638443195261
The number of items in train is:
28
The loss for epoch 0
0.5487994439983075
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/media/graph/graph_0_summary_d2013d53.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/media
The running loss is: 19.380546062253416 The number of items in train is: 28 The loss for epoch 1 0.6921623593661934 The running loss is: 13.36081693135202 The number of items in train is: 28 The loss for epoch 2 0.4771720332625721
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-summary.json
The running loss is: 11.953934513032436 The number of items in train is: 28 The loss for epoch 3 0.4269262326083013 1 The running loss is: 12.163511937658768 The number of items in train is: 28 The loss for epoch 4 0.43441114063067027 The running loss is: 9.378338906506542 The number of items in train is: 28 The loss for epoch 5 0.33494067523237653 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-summary.json
The running loss is: 12.704894061491359 The number of items in train is: 28 The loss for epoch 6 0.4537462164818343 The running loss is: 11.437989893951453 The number of items in train is: 28 The loss for epoch 7 0.4084996390696948 1 The running loss is: 11.367679933318868 The number of items in train is: 28 The loss for epoch 8 0.40598856904710245
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-summary.json
The running loss is: 11.42585400538519 The number of items in train is: 28 The loss for epoch 9 0.40806621447804253 1 Data saved to: 28_May_202005_15PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_15PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['qludt1wz'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 802.753906
66 2020-04-22 sub_region ... 66 877.084839
67 2020-04-23 sub_region ... 67 830.681274
68 2020-04-24 sub_region ... 68 816.030273
69 2020-04-25 sub_region ... 69 821.921509
70 2020-04-26 sub_region ... 70 826.182373
71 2020-04-27 sub_region ... 71 846.830322
72 2020-04-28 sub_region ... 72 856.710266
73 2020-04-29 sub_region ... 73 857.412720
74 2020-04-30 sub_region ... 74 841.012085
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/media/plotly/test_plot_20_e30096eb.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: qludt1wz
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/media/plotly INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171546-qludt1wz/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: qludt1wz INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: i3u83f7j with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 2 use_mask: True wandb: Agent Started Run: i3u83f7j
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/i3u83f7j INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmkzdTgzZjdqOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
10.473457891494036
The number of items in train is:
27
The loss for epoch 0
0.38790584783311244
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/media/graph/graph_0_summary_c269f7d4.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/media
The running loss is: 12.50141921825707 The number of items in train is: 27 The loss for epoch 1 0.4630155266021137 1 The running loss is: 12.783501834142953 The number of items in train is: 27 The loss for epoch 2 0.47346303089418346 The running loss is: 5.422599470242858 The number of items in train is: 27 The loss for epoch 3 0.20083701741640214
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-summary.json
1 The running loss is: 5.399664411786944 The number of items in train is: 27 The loss for epoch 4 0.19998757080692384 The running loss is: 4.413807349279523 The number of items in train is: 27 The loss for epoch 5 0.16347434626961196 The running loss is: 4.564219214487821 The number of items in train is: 27 The loss for epoch 6 0.16904515609214152
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-summary.json
The running loss is: 4.552914118394256 The number of items in train is: 27 The loss for epoch 7 0.16862644882941688 1 The running loss is: 4.2418790506199 The number of items in train is: 27 The loss for epoch 8 0.15710663150444074 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-summary.json
The running loss is: 4.331287162844092 The number of items in train is: 27 The loss for epoch 9 0.16041804306829968 Data saved to: 28_May_202005_16PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_16PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['i3u83f7j']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 839.469238
66 2020-04-22 sub_region ... 66 751.845520
67 2020-04-23 sub_region ... 67 796.254761
68 2020-04-24 sub_region ... 68 840.047058
69 2020-04-25 sub_region ... 69 856.757812
70 2020-04-26 sub_region ... 70 857.516724
71 2020-04-27 sub_region ... 71 805.222778
72 2020-04-28 sub_region ... 72 846.005188
73 2020-04-29 sub_region ... 73 746.322815
74 2020-04-30 sub_region ... 74 799.226501
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/media/plotly/test_plot_20_908a465b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/media/plotly
wandb: Agent Finished Run: i3u83f7j
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171601-i3u83f7j/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: i3u83f7j INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ud7enlgo with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 2 use_mask: False wandb: Agent Started Run: ud7enlgo
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ud7enlgo INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnVkN2VubGdvOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
10.56319503299892
The number of items in train is:
27
The loss for epoch 0
0.39122944566662665
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/media/graph/graph_0_summary_fb5a18e4.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/media/graph
The running loss is: 12.365171766839921 The number of items in train is: 27 The loss for epoch 1 0.45796932469777485 1 The running loss is: 10.063402196858078 The number of items in train is: 27 The loss for epoch 2 0.37271859988363254
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-summary.json
The running loss is: 5.191733076237142 The number of items in train is: 27 The loss for epoch 3 0.19228641023100526 1 The running loss is: 5.048446481116116 The number of items in train is: 27 The loss for epoch 4 0.1869794993005969 The running loss is: 4.6334264650940895 The number of items in train is: 27 The loss for epoch 5 0.17160838759607738 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-summary.json
The running loss is: 4.809368921909481 The number of items in train is: 27 The loss for epoch 6 0.17812477488553635 The running loss is: 4.469904407858849 The number of items in train is: 27 The loss for epoch 7 0.16555201510588327 1 The running loss is: 4.4312631585635245 The number of items in train is: 27 The loss for epoch 8 0.164120857724575
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-summary.json
The running loss is: 4.365964470198378 The number of items in train is: 27 The loss for epoch 9 0.1617023877851251 1 Data saved to: 28_May_202005_16PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_16PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['ud7enlgo'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 820.661011
66 2020-04-22 sub_region ... 66 731.493591
67 2020-04-23 sub_region ... 67 777.978394
68 2020-04-24 sub_region ... 68 817.230530
69 2020-04-25 sub_region ... 69 833.734131
70 2020-04-26 sub_region ... 70 830.960205
71 2020-04-27 sub_region ... 71 781.981934
72 2020-04-28 sub_region ... 72 821.894836
73 2020-04-29 sub_region ... 73 732.720215
74 2020-04-30 sub_region ... 74 782.010132
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/media/plotly/test_plot_20_a6760719.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ud7enlgo
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171612-ud7enlgo/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ud7enlgo INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: wwxi05mq with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 3 use_mask: True wandb: Agent Started Run: wwxi05mq
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/wwxi05mq INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnd3eGkwNW1xOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
22.41447990387678
The number of items in train is:
27
The loss for epoch 0
0.8301659223658068
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/media/graph/graph_0_summary_486438ab.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/media
The running loss is: 20.35543255507946 The number of items in train is: 27 The loss for epoch 1 0.7539049094473874 1 The running loss is: 19.27651271224022 The number of items in train is: 27 The loss for epoch 2 0.7139449152681563
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-summary.json
The running loss is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-history.jsonl
12.351722501218319 The number of items in train is: 27 The loss for epoch 3 0.4574712037488266 1 The running loss is: 9.813126027584076 The number of items in train is: 27 The loss for epoch 4 0.36344911213274356 The running loss is: 6.200845863670111 The number of items in train is: 27 The loss for epoch 5 0.2296609579137078 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-history.jsonl
The running loss is: 5.99026171118021 The number of items in train is: 27 The loss for epoch 6 0.2218615448585263 The running loss is: 5.51112406142056 The number of items in train is: 27 The loss for epoch 7 0.20411570597853926 1 The running loss is: 5.962472440674901 The number of items in train is: 27 The loss for epoch 8 0.22083231261758893
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-history.jsonl
The running loss is: 5.207641955465078 The number of items in train is: 27 The loss for epoch 9 0.1928756279801881 1 Data saved to: 28_May_202005_16PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['wwxi05mq']
Data saved to: 28_May_202005_16PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 731.740784
66 2020-04-22 sub_region ... 66 738.606201
67 2020-04-23 sub_region ... 67 737.777405
68 2020-04-24 sub_region ... 68 726.210022
69 2020-04-25 sub_region ... 69 719.908691
70 2020-04-26 sub_region ... 70 719.150635
71 2020-04-27 sub_region ... 71 720.189941
72 2020-04-28 sub_region ... 72 731.834473
73 2020-04-29 sub_region ... 73 735.288940
74 2020-04-30 sub_region ... 74 731.156189
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/media/plotly/test_plot_20_3857be08.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: wwxi05mq
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171622-wwxi05mq/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: wwxi05mq INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: g7arybyl with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 3 use_mask: False wandb: Agent Started Run: g7arybyl
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/g7arybyl INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmc3YXJ5YnlsOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
22.289740599691868
The number of items in train is:
27
The loss for epoch 0
0.8255459481367359
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/media/graph/graph_0_summary_fcaaf699.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/media/graph
The running loss is: 20.721937209367752 The number of items in train is: 27 The loss for epoch 1 0.7674791559025094 1 The running loss is: 19.501514554023743 The number of items in train is: 27 The loss for epoch 2 0.7222783168156942
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-history.jsonl
The running loss is: 13.022847175598145 The number of items in train is: 27 The loss for epoch 3 0.4823276731703017 1 The running loss is: 9.215275160968304 The number of items in train is: 27 The loss for epoch 4 0.3413064874432705 2 The running loss is: 6.453644126653671 The number of items in train is: 27 The loss for epoch 5 0.23902385654272856
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-history.jsonl
The running loss is: 6.3195782247930765 The number of items in train is: 27 The loss for epoch 6 0.23405845277011395 1 The running loss is: 5.538634759373963 The number of items in train is: 27 The loss for epoch 7 0.20513462071755417 The running loss is: 6.0211376789957285 The number of items in train is: 27 The loss for epoch 8 0.22300509922206402 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-history.jsonl
The running loss is: 5.4065384501591325 The number of items in train is: 27 The loss for epoch 9 0.2002421648207086 Data saved to: 28_May_202005_16PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_16PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['g7arybyl']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-summary.json
torch.Size([1, 10, 9])
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-history.jsonl
Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 723.085754
66 2020-04-22 sub_region ... 66 731.862305
67 2020-04-23 sub_region ... 67 732.741089
68 2020-04-24 sub_region ... 68 723.526733
69 2020-04-25 sub_region ... 69 715.763306
70 2020-04-26 sub_region ... 70 713.154541
71 2020-04-27 sub_region ... 71 712.077576
72 2020-04-28 sub_region ... 72 725.158813
73 2020-04-29 sub_region ... 73 729.568176
74 2020-04-30 sub_region ... 74 726.045166
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/media/plotly/test_plot_20_967a32da.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: g7arybyl
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171638-g7arybyl/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: g7arybyl INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: e6thpjz5 with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 4 use_mask: True wandb: Agent Started Run: e6thpjz5
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/e6thpjz5 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmU2dGhwano1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
17.69617347419262
The number of items in train is:
26
The loss for epoch 0
0.6806220566997161
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/media/graph/graph_0_summary_4c0ec5ce.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/media/graph
The running loss is: 12.942448504269123 The number of items in train is: 26 The loss for epoch 1 0.4977864809334278 1 The running loss is: 11.007499657571316 The number of items in train is: 26 The loss for epoch 2 0.4233653714450506 2 The running loss is: 7.731042757630348 The number of items in train is: 26 The loss for epoch 3 0.297347798370398
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-summary.json
The running loss is: 4.677496463060379 The number of items in train is: 26 The loss for epoch 4 0.1799037101177069 1 The running loss is: 3.537908681668341 The number of items in train is: 26 The loss for epoch 5 0.13607341083339775 The running loss is: 3.6070269970223308 The number of items in train is: 26 The loss for epoch 6 0.13873180757778195 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-summary.json
The running loss is: 5.469477290287614 The number of items in train is: 26 The loss for epoch 7 0.21036451116490823 2 The running loss is: 4.373710255138576 The number of items in train is: 26 The loss for epoch 8 0.16821962519763753 The running loss is: 4.385507521219552 The number of items in train is: 26 The loss for epoch 9 0.168673366200752
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-summary.json
Data saved to: 28_May_202005_16PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_16PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['e6thpjz5']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-summary.json
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 641.062012
66 2020-04-22 sub_region ... 66 649.469788
67 2020-04-23 sub_region ... 67 662.655884
68 2020-04-24 sub_region ... 68 663.000610
69 2020-04-25 sub_region ... 69 652.813110
70 2020-04-26 sub_region ... 70 647.735291
71 2020-04-27 sub_region ... 71 634.131409
72 2020-04-28 sub_region ... 72 641.305664
73 2020-04-29 sub_region ... 73 641.875549
74 2020-04-30 sub_region ... 74 656.309570
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/media/plotly/test_plot_20_7320596b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: e6thpjz5
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171655-e6thpjz5/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: e6thpjz5 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: awvxuc0v with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 4 use_mask: False wandb: Agent Started Run: awvxuc0v
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/awvxuc0v INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmF3dnh1YzB2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
17.936274334788322
The number of items in train is:
26
The loss for epoch 0
0.6898567051841662
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/media/graph/graph_0_summary_6048a156.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/media/graph
The running loss is: 12.563346169888973 The number of items in train is: 26 The loss for epoch 1 0.48320562191880667 1 The running loss is: 10.63600341975689 The number of items in train is: 26 The loss for epoch 2 0.4090770546060342 2 The running loss is: 7.032686905935407 The number of items in train is: 26 The loss for epoch 3 0.27048795792059255
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-history.jsonl
The running loss is: 4.1291360929608345 The number of items in train is: 26 The loss for epoch 4 0.1588129266523398 1 The running loss is: 3.5215289955958724 The number of items in train is: 26 The loss for epoch 5 0.13544342290753356 The running loss is: 4.55407845415175 The number of items in train is: 26 The loss for epoch 6 0.17515686362122113
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-summary.json
1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-history.jsonl
The running loss is: 3.83646054379642 The number of items in train is: 26 The loss for epoch 7 0.14755617476140076 2 The running loss is: 4.684140918310732 The number of items in train is: 26 The loss for epoch 8 0.18015926608887428 The running loss is: 7.284909620881081 The number of items in train is: 26 The loss for epoch 9 0.28018883157234925
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-history.jsonl
1 Data saved to: 28_May_202005_17PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_17PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
INFO:wandb.wandb_agent:Running runs: ['awvxuc0v']
date level ... original_index preds 55 2020-04-11 sub_region ... 55 0.000000 56 2020-04-12 sub_region ... 56 0.000000 57 2020-04-13 sub_region ... 57 0.000000 58 2020-04-14 sub_region ... 58 0.000000 59 2020-04-15 sub_region ... 59 0.000000 60 2020-04-16 sub_region ... 60 0.000000 61 2020-04-17 sub_region ... 61 0.000000 62 2020-04-18 sub_region ... 62 0.000000 63 2020-04-19 sub_region ... 63 0.000000 64 2020-04-20 sub_region ... 64 0.000000 65 2020-04-21 sub_region ... 65 811.324219 66 2020-04-22 sub_region ... 66 801.534241 67 2020-04-23 sub_region ... 67 837.117920 68 2020-04-24 sub_region ... 68 854.320984 69 2020-04-25 sub_region ... 69 834.952698 70 2020-04-26 sub_region ... 70 831.260376 71 2020-04-27 sub_region ... 71 782.874146 72 2020-04-28 sub_region ... 72 832.120972 73 2020-04-29 sub_region ... 73 785.114258 74 2020-04-30 sub_region ... 74 817.378418 [20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: awvxuc0v
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/media/plotly/test_plot_20_d2368e1b.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171710-awvxuc0v/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: awvxuc0v INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: rjmkv5tz with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 5 use_mask: True wandb: Agent Started Run: rjmkv5tz
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/rjmkv5tz INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnJqbWt2NXR6OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
16.614094384014606
The number of items in train is:
26
The loss for epoch 0
0.639003630154408
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/media/graph/graph_0_summary_5ed96a4e.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/media
The running loss is: 12.784377932548523 The number of items in train is: 26 The loss for epoch 1 0.49170684355955857 1 The running loss is: 10.317914854735136 The number of items in train is: 26 The loss for epoch 2 0.3968428790282745 The running loss is: 8.098226331174374 The number of items in train is: 26 The loss for epoch 3 0.31147024350670666 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-history.jsonl
The running loss is: 5.721522863255814 The number of items in train is: 26 The loss for epoch 4 0.22005857166368514 The running loss is: 4.689897304400802 The number of items in train is: 26 The loss for epoch 5 0.180380665553877 The running loss is: 6.495792439207435 The number of items in train is: 26 The loss for epoch 6 0.24983817073874748 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-history.jsonl
The running loss is: 5.418599771335721 The number of items in train is: 26 The loss for epoch 7 0.20840768351291233 The running loss is: 4.195066409185529 The number of items in train is: 26 The loss for epoch 8 0.16134870804559726 1 The running loss is: 5.375888114795089 The number of items in train is: 26 The loss for epoch 9 0.2067649274921188 2 Data saved to: 28_May_202005_17PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-history.jsonl
Data saved to: 28_May_202005_17PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['rjmkv5tz']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 730.216797
66 2020-04-22 sub_region ... 66 741.966797
67 2020-04-23 sub_region ... 67 738.005005
68 2020-04-24 sub_region ... 68 724.475220
69 2020-04-25 sub_region ... 69 719.175964
70 2020-04-26 sub_region ... 70 717.805908
71 2020-04-27 sub_region ... 71 714.661133
72 2020-04-28 sub_region ... 72 731.062378
73 2020-04-29 sub_region ... 73 725.214966
74 2020-04-30 sub_region ... 74 723.802612
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/media/plotly/test_plot_20_54bf2aea.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/media/plotly
wandb: Agent Finished Run: rjmkv5tz
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171721-rjmkv5tz/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: rjmkv5tz INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: xl6invys with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 1 out_seq_length: 5 use_mask: False wandb: Agent Started Run: xl6invys
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/xl6invys INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnhsNmludnlzOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
16.377284348011017
The number of items in train is:
26
The loss for epoch 0
0.6298955518465775
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/media/graph/graph_0_summary_53b91896.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/media
The running loss is: 14.166790530085564 The number of items in train is: 26 The loss for epoch 1 0.5448765588494447 The running loss is: 9.428598899394274 The number of items in train is: 26 The loss for epoch 2 0.3626384192074721 1 The running loss is: 6.6330753564834595 The number of items in train is: 26 The loss for epoch 3 0.2551182829416715 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/wandb-history.jsonl
The running loss is: 5.67338194232434 The number of items in train is: 26 The loss for epoch 4 0.2182069977817054 3 Stopping model now Data saved to: 28_May_202005_17PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_17PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 339.937469
66 2020-04-22 sub_region ... 66 348.662567
67 2020-04-23 sub_region ... 67 321.414612
68 2020-04-24 sub_region ... 68 302.783997
69 2020-04-25 sub_region ... 69 324.099854
70 2020-04-26 sub_region ... 70 317.279297
71 2020-04-27 sub_region ... 71 371.108948
72 2020-04-28 sub_region ... 72 297.625122
73 2020-04-29 sub_region ... 73 384.647125
74 2020-04-30 sub_region ... 74 332.773010
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.wandb_agent:Running runs: ['xl6invys']
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/media/plotly/test_plot_10_807ad6e3.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: xl6invys
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/wandb-history.jsonl INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/media/plotly/test_plot_all_11_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/media/plotly INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171732-xl6invys/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: xl6invys INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: iamu2dq6 with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 1 use_mask: True wandb: Agent Started Run: iamu2dq6
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/iamu2dq6 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmlhbXUyZHE2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/media/graph/graph_0_summary_a8e312dd.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/media
The running loss is: 16.114518700167537 The number of items in train is: 28 The loss for epoch 0 0.5755185250059834 The running loss is: 40.93923968449235 The number of items in train is: 28 The loss for epoch 1 1.462115703017584
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-history.jsonl
The running loss is: 17.72650630073622 The number of items in train is: 28 The loss for epoch 2 0.6330895107405793 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-history.jsonl
The running loss is: 13.82603452168405 The number of items in train is: 28 The loss for epoch 3 0.4937869472030018 The running loss is: 12.235369089990854 The number of items in train is: 28 The loss for epoch 4 0.43697746749967337
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-history.jsonl
The running loss is: 12.42129836557433 The number of items in train is: 28 The loss for epoch 5 0.4436177987705118 1
INFO:wandb.wandb_agent:Running runs: ['iamu2dq6']
The running loss is: 11.593332045944408 The number of items in train is: 28 The loss for epoch 6 0.4140475730694431
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-history.jsonl
The running loss is: 12.022986069088802 The number of items in train is: 28 The loss for epoch 7 0.42939235961031436 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-history.jsonl
The running loss is: 11.050711158488411 The number of items in train is: 28 The loss for epoch 8 0.3946682556603004 2 The running loss is: 10.95676115965398 The number of items in train is: 28 The loss for epoch 9 0.3913128985590707 3 Stopping model now Data saved to: 28_May_202005_17PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-history.jsonl
Data saved to: 28_May_202005_17PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 837.403687
66 2020-04-22 sub_region ... 66 830.898865
67 2020-04-23 sub_region ... 67 824.932678
68 2020-04-24 sub_region ... 68 836.498291
69 2020-04-25 sub_region ... 69 841.817627
70 2020-04-26 sub_region ... 70 846.289368
71 2020-04-27 sub_region ... 71 830.351929
72 2020-04-28 sub_region ... 72 852.223999
73 2020-04-29 sub_region ... 73 823.185059
74 2020-04-30 sub_region ... 74 832.774414
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: iamu2dq6
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/media/plotly/test_plot_20_754d0a08.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171742-iamu2dq6/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: iamu2dq6 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: dkl5ngk8 with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 1 use_mask: False wandb: Agent Started Run: dkl5ngk8
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/dkl5ngk8 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmRrbDVuZ2s4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/media/graph/graph_0_summary_4935b97e.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/media
The running loss is: 15.39689282909967 The number of items in train is: 28 The loss for epoch 0 0.5498890296107025
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-summary.json
The running loss is: 44.767726235091686 The number of items in train is: 28 The loss for epoch 1 1.5988473655389888 The running loss is: 22.245736518874764 The number of items in train is: 28 The loss for epoch 2 0.794490589959813 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-summary.json
The running loss is: 15.605747589841485 The number of items in train is: 28 The loss for epoch 3 0.5573481282086244 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-summary.json
The running loss is: 18.130321642383933 The number of items in train is: 28 The loss for epoch 4 0.6475114872279976 3 Stopping model now Data saved to: 28_May_202005_18PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['dkl5ngk8']
Data saved to: 28_May_202005_18PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/config.yaml
CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 410.513519
66 2020-04-22 sub_region ... 66 402.391907
67 2020-04-23 sub_region ... 67 409.347412
68 2020-04-24 sub_region ... 68 416.087708
69 2020-04-25 sub_region ... 69 416.266327
70 2020-04-26 sub_region ... 70 415.997223
71 2020-04-27 sub_region ... 71 405.508575
72 2020-04-28 sub_region ... 72 415.981689
73 2020-04-29 sub_region ... 73 393.018555
74 2020-04-30 sub_region ... 74 405.671448
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/media/plotly/test_plot_10_d4fb56b9.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: dkl5ngk8
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/media/plotly/test_plot_all_11_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171758-dkl5ngk8/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: dkl5ngk8 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: op8okjem with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 2 use_mask: True wandb: Agent Started Run: op8okjem
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/op8okjem INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm9wOG9ramVtOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/media/graph/graph_0_summary_ab59a122.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/media/graph
The running loss is: 10.718430198729038 The number of items in train is: 27 The loss for epoch 0 0.3969788962492236 The running loss is: 28.698574017733335 The number of items in train is: 27 The loss for epoch 1 1.0629101488049384
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-summary.json
The running loss is: 16.018407717347145 The number of items in train is: 27 The loss for epoch 2 0.5932743599017462 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-summary.json
The running loss is: 10.534781254827976 The number of items in train is: 27 The loss for epoch 3 0.39017708351214725 2 The running loss is: 5.385364955291152 The number of items in train is: 27 The loss for epoch 4 0.1994579613070797
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-summary.json
The running loss is: 7.625352939590812 The number of items in train is: 27 The loss for epoch 5 0.28242047924410413 1
INFO:wandb.wandb_agent:Running runs: ['op8okjem'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-summary.json
The running loss is: 6.7238668041536584 The number of items in train is: 27 The loss for epoch 6 0.2490321038575429 The running loss is: 7.8772421116009355 The number of items in train is: 27 The loss for epoch 7 0.29174970783707166
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-summary.json
The running loss is: 7.317102946341038 The number of items in train is: 27 The loss for epoch 8 0.27100381282744584 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-summary.json
The running loss is: 5.420254968106747 The number of items in train is: 27 The loss for epoch 9 0.20075018400395359 Data saved to: 28_May_202005_18PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_18PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9])
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-history.jsonl
Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-summary.json /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 779.043823
66 2020-04-22 sub_region ... 66 778.755005
67 2020-04-23 sub_region ... 67 778.881592
68 2020-04-24 sub_region ... 68 779.025024
69 2020-04-25 sub_region ... 69 779.112366
70 2020-04-26 sub_region ... 70 779.083008
71 2020-04-27 sub_region ... 71 778.987671
72 2020-04-28 sub_region ... 72 778.948608
73 2020-04-29 sub_region ... 73 778.723877
74 2020-04-30 sub_region ... 74 778.891602
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/media/plotly/test_plot_20_03adc1c1.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: op8okjem
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171814-op8okjem/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: op8okjem INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: kmhcvsx4 with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 2 use_mask: False wandb: Agent Started Run: kmhcvsx4
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/kmhcvsx4 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmttaGN2c3g0OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/media/graph/graph_0_summary_a57114ef.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/media/graph
The running loss is: 10.668751269578934 The number of items in train is: 27 The loss for epoch 0 0.3951389359103309 The running loss is: 28.521158926188946 The number of items in train is: 27 The loss for epoch 1 1.0563392194884795
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-history.jsonl
The running loss is: 14.176970317959785 The number of items in train is: 27 The loss for epoch 2 0.5250729747392513
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-history.jsonl
The running loss is: 6.591888493858278 The number of items in train is: 27 The loss for epoch 3 0.24414401829104732 1 The running loss is: 7.487370473332703 The number of items in train is: 27 The loss for epoch 4 0.27731001753084084
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-history.jsonl
The running loss is: 5.26125293970108 The number of items in train is: 27 The loss for epoch 5 0.1948612199889289
INFO:wandb.wandb_agent:Running runs: ['kmhcvsx4'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-history.jsonl
The running loss is: 6.058157247491181 The number of items in train is: 27 The loss for epoch 6 0.22437619435152523 1 The running loss is: 4.886977478861809 The number of items in train is: 27 The loss for epoch 7 0.18099916588377069
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-history.jsonl
The running loss is: 8.049873136915267 The number of items in train is: 27 The loss for epoch 8 0.29814344951538024 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-history.jsonl
The running loss is: 6.706313902512193 The number of items in train is: 27 The loss for epoch 9 0.24838199638934047 2 Data saved to: 28_May_202005_18PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_18PM_model.pth interpolate should be below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-history.jsonl
Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 755.005249
66 2020-04-22 sub_region ... 66 751.443115
67 2020-04-23 sub_region ... 67 753.844849
68 2020-04-24 sub_region ... 68 755.624390
69 2020-04-25 sub_region ... 69 755.841370
70 2020-04-26 sub_region ... 70 755.883606
71 2020-04-27 sub_region ... 71 752.995850
72 2020-04-28 sub_region ... 72 755.255066
73 2020-04-29 sub_region ... 73 751.328247
74 2020-04-30 sub_region ... 74 753.881714
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/media/plotly/test_plot_20_9c0dfd36.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: kmhcvsx4
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171832-kmhcvsx4/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: kmhcvsx4 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 4o5xdnw0 with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 3 use_mask: True wandb: Agent Started Run: 4o5xdnw0
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/4o5xdnw0 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjRvNXhkbncwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/media/graph/graph_0_summary_b776ceb7.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/media/graph
The running loss is: 23.849574618041515 The number of items in train is: 27 The loss for epoch 0 0.883317578445982 The running loss is: 20.568836465477943 The number of items in train is: 27 The loss for epoch 1 0.7618087579806646
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-history.jsonl
The running loss is: 19.27038937062025 The number of items in train is: 27 The loss for epoch 2 0.7137181248377871
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-history.jsonl
The running loss is: 17.25364562869072 The number of items in train is: 27 The loss for epoch 3 0.6390239121737303 1 The running loss is: 14.78194186091423 The number of items in train is: 27 The loss for epoch 4 0.5474793281820085
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-history.jsonl
The running loss is: 10.99204146116972 The number of items in train is: 27 The loss for epoch 5 0.40711264670998965 1
INFO:wandb.wandb_agent:Running runs: ['4o5xdnw0'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-history.jsonl
The running loss is: 34.4778664149344 The number of items in train is: 27 The loss for epoch 6 1.2769580153679405 The running loss is: 18.340780273079872 The number of items in train is: 27 The loss for epoch 7 0.6792881582622174 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-history.jsonl
The running loss is: 18.76011623442173 The number of items in train is: 27 The loss for epoch 8 0.6948191197933974 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-history.jsonl
The running loss is: 18.743604812771082 The number of items in train is: 27 The loss for epoch 9 0.6942075856581882 Data saved to: 28_May_202005_18PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_18PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 351.287323
66 2020-04-22 sub_region ... 66 351.352722
67 2020-04-23 sub_region ... 67 351.261353
68 2020-04-24 sub_region ... 68 351.227020
69 2020-04-25 sub_region ... 69 351.256317
70 2020-04-26 sub_region ... 70 351.258667
71 2020-04-27 sub_region ... 71 351.388000
72 2020-04-28 sub_region ... 72 351.195374
73 2020-04-29 sub_region ... 73 351.413818
74 2020-04-30 sub_region ... 74 351.286987
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/media/plotly/test_plot_20_3234c249.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 4o5xdnw0
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171848-4o5xdnw0/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 4o5xdnw0 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: iyc8jxdt with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 3 use_mask: False wandb: Agent Started Run: iyc8jxdt
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/iyc8jxdt INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOml5YzhqeGR0OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/media/graph/graph_0_summary_e1e54220.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/media
The running loss is: 23.785254307091236 The number of items in train is: 27 The loss for epoch 0 0.8809353447070828 The running loss is: 20.365187123417854 The number of items in train is: 27 The loss for epoch 1 0.7542661897562168
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-history.jsonl
The running loss is: 19.407820589840412 The number of items in train is: 27 The loss for epoch 2 0.7188081699940894
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-history.jsonl
The running loss is: 16.047366067767143 The number of items in train is: 27 The loss for epoch 3 0.594346891398783 1 The running loss is: 16.049381844699383 The number of items in train is: 27 The loss for epoch 4 0.5944215498036809
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-history.jsonl
The running loss is: 8.695411691442132 The number of items in train is: 27 The loss for epoch 5 0.3220522848682271 1
INFO:wandb.wandb_agent:Running runs: ['iyc8jxdt'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-history.jsonl
The running loss is: 11.989098764955997 The number of items in train is: 27 The loss for epoch 6 0.4440406949983703 The running loss is: 12.975145380944014 The number of items in train is: 27 The loss for epoch 7 0.48056094003496347 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-history.jsonl
The running loss is: 6.981494817882776 The number of items in train is: 27 The loss for epoch 8 0.25857388214380655
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-history.jsonl
The running loss is: 7.491051498800516 The number of items in train is: 27 The loss for epoch 9 0.27744635180742655 1 Data saved to: 28_May_202005_19PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_19PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 709.696960
66 2020-04-22 sub_region ... 66 709.161499
67 2020-04-23 sub_region ... 67 708.643188
68 2020-04-24 sub_region ... 68 708.597656
69 2020-04-25 sub_region ... 69 708.908020
70 2020-04-26 sub_region ... 70 708.755493
71 2020-04-27 sub_region ... 71 709.836182
72 2020-04-28 sub_region ... 72 708.476990
73 2020-04-29 sub_region ... 73 709.191528
74 2020-04-30 sub_region ... 74 708.709473
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/media/plotly/test_plot_20_509b02c0.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: iyc8jxdt
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171903-iyc8jxdt/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: iyc8jxdt INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ybklo6n3 with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 4 use_mask: True wandb: Agent Started Run: ybklo6n3
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ybklo6n3 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnlia2xvNm4zOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/media/graph/graph_0_summary_36165fc0.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/media/graph
The running loss is: 25.31017677485943 The number of items in train is: 26 The loss for epoch 0 0.9734683374945934 The running loss is: 17.691114902496338 The number of items in train is: 26 The loss for epoch 1 0.6804274962498591 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-summary.json
The running loss is: 21.284134477376938 The number of items in train is: 26 The loss for epoch 2 0.8186205568221899
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-summary.json
The running loss is: 14.811767756938934 The number of items in train is: 26 The loss for epoch 3 0.5696833752668821 The running loss is: 12.015214286744595 The number of items in train is: 26 The loss for epoch 4 0.46212362641325366 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-summary.json
The running loss is: 10.353260762989521 The number of items in train is: 26 The loss for epoch 5 0.3982023370380585 2
INFO:wandb.wandb_agent:Running runs: ['ybklo6n3']
The running loss is: 7.667264841496944 The number of items in train is: 26 The loss for epoch 6 0.2948948015960363
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-summary.json
3 Stopping model now Data saved to: 28_May_202005_19PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_19PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-summary.json /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 318.850616
66 2020-04-22 sub_region ... 66 316.706177
67 2020-04-23 sub_region ... 67 320.013977
68 2020-04-24 sub_region ... 68 321.565765
69 2020-04-25 sub_region ... 69 321.352051
70 2020-04-26 sub_region ... 70 320.382446
71 2020-04-27 sub_region ... 71 318.602081
72 2020-04-28 sub_region ... 72 318.317017
73 2020-04-29 sub_region ... 73 318.346680
74 2020-04-30 sub_region ... 74 320.484741
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/media/plotly/test_plot_14_b1c54b64.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ybklo6n3
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/media/plotly/test_plot_all_15_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171919-ybklo6n3/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ybklo6n3 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 5ca6mu99 with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 4 use_mask: False wandb: Agent Started Run: 5ca6mu99
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/5ca6mu99 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjVjYTZtdTk5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/media/graph/graph_0_summary_b0c413d6.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/media
The running loss is: 25.090075597167015 The number of items in train is: 26 The loss for epoch 0 0.9650029075833467 The running loss is: 17.45224541425705 The number of items in train is: 26 The loss for epoch 1 0.6712402082406558 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-history.jsonl
The running loss is: 21.34785506129265 The number of items in train is: 26 The loss for epoch 2 0.8210713485112557
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-history.jsonl
The running loss is: 14.617094554007053 The number of items in train is: 26 The loss for epoch 3 0.5621959443848866 1 The running loss is: 12.857044279575348 The number of items in train is: 26 The loss for epoch 4 0.4945017030605903
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-history.jsonl
The running loss is: 10.908861950039864 The number of items in train is: 26 The loss for epoch 5 0.4195716134630717 1
INFO:wandb.wandb_agent:Running runs: ['5ca6mu99']
The running loss is: 7.671748608350754 The number of items in train is: 26 The loss for epoch 6 0.2950672541673367
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-history.jsonl
The running loss is: 6.437005430459976 The number of items in train is: 26 The loss for epoch 7 0.2475771319407683
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-history.jsonl
The running loss is: 7.865247845649719 The number of items in train is: 26 The loss for epoch 8 0.3025095325249892 1 The running loss is: 5.2042678743600845 The number of items in train is: 26 The loss for epoch 9 0.2001641490138494 Data saved to: 28_May_202005_19PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-history.jsonl
Data saved to: 28_May_202005_19PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 570.512756
66 2020-04-22 sub_region ... 66 571.143372
67 2020-04-23 sub_region ... 67 570.903015
68 2020-04-24 sub_region ... 68 570.691162
69 2020-04-25 sub_region ... 69 570.595093
70 2020-04-26 sub_region ... 70 570.591003
71 2020-04-27 sub_region ... 71 570.901611
72 2020-04-28 sub_region ... 72 570.663696
73 2020-04-29 sub_region ... 73 571.259827
74 2020-04-30 sub_region ... 74 571.026001
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/media/plotly/test_plot_20_96e49c04.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/media/plotly
wandb: Agent Finished Run: 5ca6mu99
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171937-5ca6mu99/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 5ca6mu99 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: w8uacmny with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 5 use_mask: True wandb: Agent Started Run: w8uacmny
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/w8uacmny INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnc4dWFjbW55OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/media/graph/graph_0_summary_96d1a8d8.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/media/graph
The running loss is: 24.998789221048355 The number of items in train is: 26 The loss for epoch 0 0.9614918931172445 The running loss is: 19.62556444108486 The number of items in train is: 26 The loss for epoch 1 0.754829401580187 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-summary.json
The running loss is: 19.021592140197754 The number of items in train is: 26 The loss for epoch 2 0.7315996976999136
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-summary.json
The running loss is: 13.53062754869461 The number of items in train is: 26 The loss for epoch 3 0.5204087518728696 The running loss is: 11.939093425869942 The number of items in train is: 26 The loss for epoch 4 0.45919590099499774
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-summary.json
The running loss is: 12.370361641049385 The number of items in train is: 26 The loss for epoch 5 0.475783140040361 1
INFO:wandb.wandb_agent:Running runs: ['w8uacmny']
The running loss is: 10.105389013886452 The number of items in train is: 26 The loss for epoch 6 0.388668808226402
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-summary.json
The running loss is: 5.374523870646954 The number of items in train is: 26 The loss for epoch 7 0.20671245656334436 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-summary.json
The running loss is: 4.875691753812134 The number of items in train is: 26 The loss for epoch 8 0.18752660591585132 The running loss is: 7.123469600221142 The number of items in train is: 26 The loss for epoch 9 0.27397960000850546 1 Data saved to: 28_May_202005_20PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-summary.json
Data saved to: 28_May_202005_20PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 697.734924
66 2020-04-22 sub_region ... 66 699.246887
67 2020-04-23 sub_region ... 67 698.447754
68 2020-04-24 sub_region ... 68 697.592102
69 2020-04-25 sub_region ... 69 697.349731
70 2020-04-26 sub_region ... 70 697.215332
71 2020-04-27 sub_region ... 71 698.537048
72 2020-04-28 sub_region ... 72 697.177002
73 2020-04-29 sub_region ... 73 699.459961
74 2020-04-30 sub_region ... 74 698.878723
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/media/plotly/test_plot_20_70b23860.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: w8uacmny
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_171953-w8uacmny/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: w8uacmny INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: rv3jho0p with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 2 out_seq_length: 5 use_mask: False wandb: Agent Started Run: rv3jho0p
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/rv3jho0p INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnJ2M2pobzBwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/media/graph/graph_0_summary_7995e82c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/media/graph
The running loss is: 24.827137678861618 The number of items in train is: 26 The loss for epoch 0 0.9548899107254468 The running loss is: 19.719864428043365 The number of items in train is: 26 The loss for epoch 1 0.758456324155514 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-summary.json
The running loss is: 18.63474379479885 The number of items in train is: 26 The loss for epoch 2 0.7167209151845712
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-summary.json
The running loss is: 13.806637480854988 The number of items in train is: 26 The loss for epoch 3 0.5310245184944227 1 The running loss is: 11.331828974187374 The number of items in train is: 26 The loss for epoch 4 0.4358395759302836
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-summary.json
The running loss is: 10.775343157351017 The number of items in train is: 26 The loss for epoch 5 0.4144362752827314 1
INFO:wandb.wandb_agent:Running runs: ['rv3jho0p']
The running loss is: 10.32153557986021 The number of items in train is: 26 The loss for epoch 6 0.3969821376869312 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-summary.json
The running loss is: 7.037654765415937 The number of items in train is: 26 The loss for epoch 7 0.2706790294390745
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-summary.json
The running loss is: 14.718151992186904 The number of items in train is: 26 The loss for epoch 8 0.5660827689302655 1 The running loss is: 5.737017912324518 The number of items in train is: 26 The loss for epoch 9 0.22065453508940452 Data saved to: 28_May_202005_20PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-summary.json
Data saved to: 28_May_202005_20PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 632.474121
66 2020-04-22 sub_region ... 66 634.329102
67 2020-04-23 sub_region ... 67 632.773071
68 2020-04-24 sub_region ... 68 631.713074
69 2020-04-25 sub_region ... 69 631.196228
70 2020-04-26 sub_region ... 70 631.151917
71 2020-04-27 sub_region ... 71 632.160889
72 2020-04-28 sub_region ... 72 631.541016
73 2020-04-29 sub_region ... 73 633.977478
74 2020-04-30 sub_region ... 74 632.605713
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/media/plotly/test_plot_20_a4b0535b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/media/plotly
wandb: Agent Finished Run: rv3jho0p
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172009-rv3jho0p/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: rv3jho0p INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: yezsyz5y with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 1 use_mask: True wandb: Agent Started Run: yezsyz5y
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/yezsyz5y INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnllenN5ejV5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/media/graph/graph_0_summary_a3f207a1.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/media
The running loss is: 16.056726850569248 The number of items in train is: 28 The loss for epoch 0 0.5734545303774732
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-history.jsonl
The running loss is: 46.208357490599155 The number of items in train is: 28 The loss for epoch 1 1.6502984818071127
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-history.jsonl
The running loss is: 29.032308392226696 The number of items in train is: 28 The loss for epoch 2 1.0368681568652391
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-history.jsonl
The running loss is: 28.841471418738365 The number of items in train is: 28 The loss for epoch 3 1.0300525506692273
INFO:wandb.wandb_agent:Running runs: ['yezsyz5y'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-history.jsonl
The running loss is: 28.338371301069856 The number of items in train is: 28 The loss for epoch 4 1.0120846893239235 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-history.jsonl
The running loss is: 28.69572694809176 The number of items in train is: 28 The loss for epoch 5 1.0248473910032772 2 The running loss is: 19.646056853234768 The number of items in train is: 28 The loss for epoch 6 0.7016448876155275
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-history.jsonl
3 Stopping model now Data saved to: 28_May_202005_20PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_20PM_model.pth interpolate should be below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-summary.json
Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-history.jsonl
CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 429.154022
66 2020-04-22 sub_region ... 66 428.729492
67 2020-04-23 sub_region ... 67 429.150635
68 2020-04-24 sub_region ... 68 429.542450
69 2020-04-25 sub_region ... 69 429.516174
70 2020-04-26 sub_region ... 70 429.553772
71 2020-04-27 sub_region ... 71 428.790649
72 2020-04-28 sub_region ... 72 429.774963
73 2020-04-29 sub_region ... 73 428.176636
74 2020-04-30 sub_region ... 74 428.942596
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/media/plotly/test_plot_14_bdfadf79.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: yezsyz5y
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/media/plotly/test_plot_all_15_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172025-yezsyz5y/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: yezsyz5y INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: lwpzskh5 with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 1 use_mask: False wandb: Agent Started Run: lwpzskh5
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/lwpzskh5 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmx3cHpza2g1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/media/graph/graph_0_summary_7cbaf016.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/media
The running loss is: 13.879746047779918 The number of items in train is: 28 The loss for epoch 0 0.4957052159921399
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-history.jsonl
The running loss is: 50.16468261182308 The number of items in train is: 28 The loss for epoch 1 1.7915958075651102
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-history.jsonl
The running loss is: 28.752860818058252 The number of items in train is: 28 The loss for epoch 2 1.0268878863592232
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-history.jsonl
The running loss is: 28.75844579562545 The number of items in train is: 28 The loss for epoch 3 1.027087349843766
INFO:wandb.wandb_agent:Running runs: ['lwpzskh5'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-history.jsonl
The running loss is: 28.228818399831653 The number of items in train is: 28 The loss for epoch 4 1.0081720857082732 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-history.jsonl
The running loss is: 28.275303000351414 The number of items in train is: 28 The loss for epoch 5 1.0098322500125505 2 The running loss is: 19.328091056086123 The number of items in train is: 28 The loss for epoch 6 0.6902889662887901
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-history.jsonl
3 Stopping model now Data saved to: 28_May_202005_20PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_20PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 429.017181
66 2020-04-22 sub_region ... 66 428.986389
67 2020-04-23 sub_region ... 67 429.089111
68 2020-04-24 sub_region ... 68 429.224731
69 2020-04-25 sub_region ... 69 429.164246
70 2020-04-26 sub_region ... 70 429.191895
71 2020-04-27 sub_region ... 71 428.905151
72 2020-04-28 sub_region ... 72 429.181915
73 2020-04-29 sub_region ... 73 428.645477
74 2020-04-30 sub_region ... 74 428.955353
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/media/plotly/test_plot_14_562144b7.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: lwpzskh5
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/media/plotly/test_plot_all_15_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172041-lwpzskh5/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: lwpzskh5 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: s7nal4zt with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 2 use_mask: True wandb: Agent Started Run: s7nal4zt
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/s7nal4zt INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnM3bmFsNHp0OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/media/graph/graph_0_summary_29740593.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/media/graph
The running loss is: 9.515796359628439 The number of items in train is: 27 The loss for epoch 0 0.3524369022084607
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-summary.json
The running loss is: 31.958917625248432 The number of items in train is: 27 The loss for epoch 1 1.1836636157499418
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-summary.json
The running loss is: 17.400996170938015 The number of items in train is: 27 The loss for epoch 2 0.6444813396643709 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-summary.json
The running loss is: 17.490230657160282 The number of items in train is: 27 The loss for epoch 3 0.647786320635566
INFO:wandb.wandb_agent:Running runs: ['s7nal4zt'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-summary.json
The running loss is: 17.88597560673952 The number of items in train is: 27 The loss for epoch 4 0.6624435409903526 The running loss is: 17.275350011885166 The number of items in train is: 27 The loss for epoch 5 0.6398277782179691
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-summary.json
The running loss is: 17.80729231238365 The number of items in train is: 27 The loss for epoch 6 0.6595293449030982
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-summary.json
The running loss is: 17.339848961681128 The number of items in train is: 27 The loss for epoch 7 0.6422166282104121
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-summary.json
The running loss is: 16.749763071537018 The number of items in train is: 27 The loss for epoch 8 0.6203615952421118
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-summary.json
The running loss is: 17.160026656463742 The number of items in train is: 27 The loss for epoch 9 0.6355565428319905 Data saved to: 28_May_202005_21PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/config.yaml
28_May_202005_21PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 452.964355
66 2020-04-22 sub_region ... 66 448.373505
67 2020-04-23 sub_region ... 67 451.731903
68 2020-04-24 sub_region ... 68 454.205078
69 2020-04-25 sub_region ... 69 454.852173
70 2020-04-26 sub_region ... 70 454.945068
71 2020-04-27 sub_region ... 71 451.557831
72 2020-04-28 sub_region ... 72 453.470245
73 2020-04-29 sub_region ... 73 448.514465
74 2020-04-30 sub_region ... 74 451.503876
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/media/plotly/test_plot_20_da2b038c.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: s7nal4zt
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172058-s7nal4zt/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: s7nal4zt INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: qnt4stp8 with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 2 use_mask: False wandb: Agent Started Run: qnt4stp8
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/qnt4stp8 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnFudDRzdHA4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/media/graph/graph_0_summary_70f484d6.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/media
The running loss is: 9.540501032024622 The number of items in train is: 27 The loss for epoch 0 0.353351890074986
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-history.jsonl
The running loss is: 32.01947431452572 The number of items in train is: 27 The loss for epoch 1 1.1859064560935453
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-history.jsonl
The running loss is: 17.433912321925163 The number of items in train is: 27 The loss for epoch 2 0.6457004563675987 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-history.jsonl
The running loss is: 17.372653737664223 The number of items in train is: 27 The loss for epoch 3 0.6434316199134897
INFO:wandb.wandb_agent:Running runs: ['qnt4stp8'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-history.jsonl
The running loss is: 17.937954049557447 The number of items in train is: 27 The loss for epoch 4 0.6643686685021277 The running loss is: 17.54516338557005 The number of items in train is: 27 The loss for epoch 5 0.649820866132224
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-history.jsonl
The running loss is: 17.718922704458237 The number of items in train is: 27 The loss for epoch 6 0.6562563964614162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-history.jsonl
The running loss is: 17.30660080164671 The number of items in train is: 27 The loss for epoch 7 0.6409852148758041
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-history.jsonl
The running loss is: 16.718926943838596 The number of items in train is: 27 The loss for epoch 8 0.6192195164384665
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-history.jsonl
The running loss is: 16.827449632808566 The number of items in train is: 27 The loss for epoch 9 0.6232388752892062 Data saved to: 28_May_202005_21PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-history.jsonl
Data saved to: 28_May_202005_21PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 457.992615
66 2020-04-22 sub_region ... 66 454.040710
67 2020-04-23 sub_region ... 67 456.812286
68 2020-04-24 sub_region ... 68 459.140167
69 2020-04-25 sub_region ... 69 459.434723
70 2020-04-26 sub_region ... 70 459.551544
71 2020-04-27 sub_region ... 71 456.045074
72 2020-04-28 sub_region ... 72 458.655548
73 2020-04-29 sub_region ... 73 453.452576
74 2020-04-30 sub_region ... 74 456.783661
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: qnt4stp8
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/media/plotly/test_plot_20_c566c54f.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172119-qnt4stp8/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: qnt4stp8 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: tau1uhlr with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 3 use_mask: True wandb: Agent Started Run: tau1uhlr
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/tau1uhlr INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnRhdTF1aGxyOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/media/graph/graph_0_summary_d503964e.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/media
The running loss is: 31.62208504229784 The number of items in train is: 27 The loss for epoch 0 1.17118833489992
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-history.jsonl
The running loss is: 19.408665716648102 The number of items in train is: 27 The loss for epoch 1 0.7188394709869668 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-history.jsonl
The running loss is: 20.407186299562454 The number of items in train is: 27 The loss for epoch 2 0.7558217147986094
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-history.jsonl
The running loss is: 20.506580412387848 The number of items in train is: 27 The loss for epoch 3 0.759502978236587
INFO:wandb.wandb_agent:Running runs: ['tau1uhlr'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-history.jsonl
The running loss is: 20.30820331722498 The number of items in train is: 27 The loss for epoch 4 0.7521556784157399 The running loss is: 19.309663198888302 The number of items in train is: 27 The loss for epoch 5 0.7151727110699371
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-history.jsonl
1 The running loss is: 18.341759957373142 The number of items in train is: 27 The loss for epoch 6 0.679324442865672
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-history.jsonl
The running loss is: 17.737579837441444 The number of items in train is: 27 The loss for epoch 7 0.6569474013867201 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-history.jsonl
The running loss is: 19.28192499279976 The number of items in train is: 27 The loss for epoch 8 0.7141453701036947
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-history.jsonl
The running loss is: 18.758581429719925 The number of items in train is: 27 The loss for epoch 9 0.694762275174812 1 Data saved to: 28_May_202005_21PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_21PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 337.341278
66 2020-04-22 sub_region ... 66 337.345581
67 2020-04-23 sub_region ... 67 337.355377
68 2020-04-24 sub_region ... 68 337.360901
69 2020-04-25 sub_region ... 69 337.360565
70 2020-04-26 sub_region ... 70 337.355988
71 2020-04-27 sub_region ... 71 337.354065
72 2020-04-28 sub_region ... 72 337.343140
73 2020-04-29 sub_region ... 73 337.349548
74 2020-04-30 sub_region ... 74 337.355652
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: tau1uhlr
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/media/plotly/test_plot_20_e91edfe1.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172140-tau1uhlr/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: tau1uhlr INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 67mx7y58 with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 3 use_mask: False wandb: Agent Started Run: 67mx7y58
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/67mx7y58 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjY3bXg3eTU4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/media/graph/graph_0_summary_7b6f7b4e.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/media
The running loss is: 31.686953715980053 The number of items in train is: 27 The loss for epoch 0 1.1735908783696316
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-history.jsonl
The running loss is: 19.326208144426346 The number of items in train is: 27 The loss for epoch 1 0.7157854868306054 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-history.jsonl
The running loss is: 20.44809077680111 The number of items in train is: 27 The loss for epoch 2 0.7573366954370782
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-history.jsonl
The running loss is: 20.70396327972412 The number of items in train is: 27 The loss for epoch 3 0.7668134548045971
INFO:wandb.wandb_agent:Running runs: ['67mx7y58'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-history.jsonl
The running loss is: 20.114036343991756 The number of items in train is: 27 The loss for epoch 4 0.7449643090367317 1 The running loss is: 19.89861784130335 The number of items in train is: 27 The loss for epoch 5 0.7369858459741981
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-history.jsonl
The running loss is: 18.967432893812656 The number of items in train is: 27 The loss for epoch 6 0.7024975145856539 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-history.jsonl
The running loss is: 19.17745415121317 The number of items in train is: 27 The loss for epoch 7 0.7102760796745619
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-history.jsonl
The running loss is: 19.25604397803545 The number of items in train is: 27 The loss for epoch 8 0.713186814001313
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-history.jsonl
The running loss is: 18.80481818318367 The number of items in train is: 27 The loss for epoch 9 0.6964747475253211 Data saved to: 28_May_202005_22PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_22PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 353.958618
66 2020-04-22 sub_region ... 66 353.994934
67 2020-04-23 sub_region ... 67 353.966064
68 2020-04-24 sub_region ... 68 353.948608
69 2020-04-25 sub_region ... 69 353.950348
70 2020-04-26 sub_region ... 70 353.951172
71 2020-04-27 sub_region ... 71 353.985535
72 2020-04-28 sub_region ... 72 353.945618
73 2020-04-29 sub_region ... 73 354.007202
74 2020-04-30 sub_region ... 74 353.972107
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/media/plotly/test_plot_20_384740d4.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 67mx7y58
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172201-67mx7y58/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 67mx7y58 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: pxdi511t with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 4 use_mask: True wandb: Agent Started Run: pxdi511t
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/pxdi511t INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnB4ZGk1MTF0OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/media/graph/graph_0_summary_0b78d094.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/media
The running loss is: 28.677166178822517 The number of items in train is: 26 The loss for epoch 0 1.1029679299547122
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-history.jsonl
The running loss is: 19.185990750789642 The number of items in train is: 26 The loss for epoch 1 0.737922721184217
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-history.jsonl
The running loss is: 20.213715583086014 The number of items in train is: 26 The loss for epoch 2 0.777450599349462
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-history.jsonl
The running loss is: 15.579263865947723 The number of items in train is: 26 The loss for epoch 3 0.5992024563826047 1
INFO:wandb.wandb_agent:Running runs: ['pxdi511t']
The running loss is: 15.614615187048912 The number of items in train is: 26 The loss for epoch 4 0.6005621225788043
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-history.jsonl
The running loss is: 14.192182525992393 The number of items in train is: 26 The loss for epoch 5 0.5458531740766305 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-history.jsonl
The running loss is: 17.893233135342598 The number of items in train is: 26 The loss for epoch 6 0.6882012744362538
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-history.jsonl
The running loss is: 15.887526050209999 The number of items in train is: 26 The loss for epoch 7 0.6110586942388461
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-history.jsonl
The running loss is: 13.056910991668701 The number of items in train is: 26 The loss for epoch 8 0.50218888429495 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-history.jsonl
The running loss is: 10.211280450224876 The number of items in train is: 26 The loss for epoch 9 0.39274155577787984 2 Data saved to: 28_May_202005_22PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_22PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 409.355835
66 2020-04-22 sub_region ... 66 409.509674
67 2020-04-23 sub_region ... 67 409.471558
68 2020-04-24 sub_region ... 68 409.399261
69 2020-04-25 sub_region ... 69 409.354004
70 2020-04-26 sub_region ... 70 409.365173
71 2020-04-27 sub_region ... 71 409.424011
72 2020-04-28 sub_region ... 72 409.376404
73 2020-04-29 sub_region ... 73 409.552612
74 2020-04-30 sub_region ... 74 409.473022
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/media/plotly/test_plot_20_eefb9a0f.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: pxdi511t
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172222-pxdi511t/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: pxdi511t INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 7f4kxu00 with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 4 use_mask: False wandb: Agent Started Run: 7f4kxu00
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/7f4kxu00 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjdmNGt4dTAwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/media/graph/graph_0_summary_a08b7baa.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/media/graph
The running loss is: 28.61015349626541 The number of items in train is: 26 The loss for epoch 0 1.1003905190871313
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-history.jsonl
The running loss is: 19.129196166992188 The number of items in train is: 26 The loss for epoch 1 0.7357383141150842
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-history.jsonl
The running loss is: 20.223093792796135 The number of items in train is: 26 The loss for epoch 2 0.7778112997229283
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-history.jsonl
The running loss is: 15.592137679457664 The number of items in train is: 26 The loss for epoch 3 0.599697603056064 1
INFO:wandb.wandb_agent:Running runs: ['7f4kxu00']
The running loss is: 14.741975575685501 The number of items in train is: 26 The loss for epoch 4 0.5669990606032885
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-history.jsonl
The running loss is: 14.141681730747223 The number of items in train is: 26 The loss for epoch 5 0.5439108357979701 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-history.jsonl
The running loss is: 17.65192748606205 The number of items in train is: 26 The loss for epoch 6 0.6789202879254634
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-history.jsonl
The running loss is: 13.934842199087143 The number of items in train is: 26 The loss for epoch 7 0.5359554691956594 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-history.jsonl
The running loss is: 10.86246071755886 The number of items in train is: 26 The loss for epoch 8 0.4177869506753408
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-history.jsonl
The running loss is: 9.102634258568287 The number of items in train is: 26 The loss for epoch 9 0.3501013176372418 1 Data saved to: 28_May_202005_22PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_22PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 475.105896
66 2020-04-22 sub_region ... 66 475.061249
67 2020-04-23 sub_region ... 67 475.085632
68 2020-04-24 sub_region ... 68 475.112213
69 2020-04-25 sub_region ... 69 475.114807
70 2020-04-26 sub_region ... 70 475.139618
71 2020-04-27 sub_region ... 71 475.079346
72 2020-04-28 sub_region ... 72 475.173126
73 2020-04-29 sub_region ... 73 475.043427
74 2020-04-30 sub_region ... 74 475.085358
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/media/plotly/test_plot_20_114ca43c.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 7f4kxu00
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172243-7f4kxu00/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 7f4kxu00 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: yobxmwpd with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 5 use_mask: True wandb: Agent Started Run: yobxmwpd
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/yobxmwpd INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnlvYnhtd3BkOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/media/graph/graph_0_summary_68b4e924.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/media/graph
The running loss is: 26.287698671221733 The number of items in train is: 26 The loss for epoch 0 1.0110653335085282
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-history.jsonl
The running loss is: 22.019511476159096 The number of items in train is: 26 The loss for epoch 1 0.8469042875445806
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-history.jsonl
The running loss is: 19.7527959048748 The number of items in train is: 26 The loss for epoch 2 0.7597229194182616
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-history.jsonl
The running loss is: 15.987582370638847 The number of items in train is: 26 The loss for epoch 3 0.6149070142553403 1
INFO:wandb.wandb_agent:Running runs: ['yobxmwpd']
The running loss is: 15.31729331612587 The number of items in train is: 26
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-history.jsonl
The loss for epoch 4 0.5891266660048411 The running loss is: 13.650041669607162 The number of items in train is: 26 The loss for epoch 5 0.5250016026771985 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-history.jsonl
The running loss is: 13.348268389701843 The number of items in train is: 26 The loss for epoch 6 0.5133949380654556 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-history.jsonl
The running loss is: 11.254892647266388 The number of items in train is: 26 The loss for epoch 7 0.4328804864333226 3 Stopping model now Data saved to: 28_May_202005_23PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_23PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 347.929901
66 2020-04-22 sub_region ... 66 348.376862
67 2020-04-23 sub_region ... 67 348.171692
68 2020-04-24 sub_region ... 68 347.880554
69 2020-04-25 sub_region ... 69 347.761993
70 2020-04-26 sub_region ... 70 347.774658
71 2020-04-27 sub_region ... 71 348.046265
72 2020-04-28 sub_region ... 72 347.813934
73 2020-04-29 sub_region ... 73 348.664154
74 2020-04-30 sub_region ... 74 348.247253
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/media/plotly/test_plot_16_3f181331.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: yobxmwpd
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/media/plotly/test_plot_all_17_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172304-yobxmwpd/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: yobxmwpd INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: tuuqqbru with config: batch_size: 2 forecast_history: 10 lr: 0.002 number_encoder_layers: 3 out_seq_length: 5 use_mask: False wandb: Agent Started Run: tuuqqbru
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/tuuqqbru INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnR1dXFxYnJ1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/media/graph/graph_0_summary_9177faf0.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/media/graph
The running loss is: 26.332314282655716 The number of items in train is: 26 The loss for epoch 0 1.0127813185636814
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-history.jsonl
The running loss is: 21.822422578930855 The number of items in train is: 26 The loss for epoch 1 0.8393239453434944
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-history.jsonl
The running loss is: 19.969776809215546 The number of items in train is: 26 The loss for epoch 2 0.7680683388159826
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-history.jsonl
The running loss is: 15.921830900013447 The number of items in train is: 26 The loss for epoch 3 0.6123781115389787 1
INFO:wandb.wandb_agent:Running runs: ['tuuqqbru']
The running loss is: 15.494909837841988 The number of items in train is: 26 The loss for epoch 4 0.5959580706862303
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-history.jsonl
The running loss is: 14.208502128720284 The number of items in train is: 26 The loss for epoch 5 0.5464808511046263 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-history.jsonl
The running loss is: 16.151983082294464 The number of items in train is: 26 The loss for epoch 6 0.621230118549787 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-history.jsonl
The running loss is: 12.41115165501833 The number of items in train is: 26 The loss for epoch 7 0.4773519867314742
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-history.jsonl
The running loss is: 9.447976488620043 The number of items in train is: 26 The loss for epoch 8 0.3633837111007709 The running loss is: 9.989659074693918 The number of items in train is: 26 The loss for epoch 9 0.3842176567189969
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-history.jsonl
1 Data saved to: 28_May_202005_23PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_23PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-history.jsonl /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/config.yaml /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 578.508301
66 2020-04-22 sub_region ... 66 578.518188
67 2020-04-23 sub_region ... 67 578.516479
68 2020-04-24 sub_region ... 68 578.510986
69 2020-04-25 sub_region ... 69 578.503906
70 2020-04-26 sub_region ... 70 578.503906
71 2020-04-27 sub_region ... 71 578.500427
72 2020-04-28 sub_region ... 72 578.506470
73 2020-04-29 sub_region ... 73 578.516052
74 2020-04-30 sub_region ... 74 578.512817
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/media/plotly/test_plot_20_ae1f0e4f.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: tuuqqbru
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172320-tuuqqbru/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: tuuqqbru INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: dm02dl9g with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 1 use_mask: True wandb: Agent Started Run: dm02dl9g
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/dm02dl9g INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmRtMDJkbDlnOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
15.545928395004012
The number of items in train is:
28
The loss for epoch 0
0.5552117283930004
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/media/graph/graph_0_summary_b9250aaf.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/media/graph
The running loss is: 21.661523299873807 The number of items in train is: 28 The loss for epoch 1 0.7736258321383502 The running loss is: 14.07579830638133 The number of items in train is: 28 The loss for epoch 2 0.5027070823707618
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-history.jsonl
The running loss is: 11.572435699170455 The number of items in train is: 28 The loss for epoch 3 0.4133012749703734 1 The running loss is: 11.783770642941818 The number of items in train is: 28 The loss for epoch 4 0.42084895153363633 The running loss is: 9.61011728970334 The number of items in train is: 28 The loss for epoch 5 0.3432184746322621 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-history.jsonl
The running loss is: 13.171849727863446 The number of items in train is: 28 The loss for epoch 6 0.4704232045665516 The running loss is: 9.562567624612711 The number of items in train is: 28 The loss for epoch 7 0.34152027230759685 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-summary.json
The running loss is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-history.jsonl
11.004484111152124 The number of items in train is: 28 The loss for epoch 8 0.3930172896840044 2 The running loss is: 10.858391232584836 The number of items in train is: 28 The loss for epoch 9 0.38779968687802985 Data saved to: 28_May_202005_23PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['dm02dl9g']
Data saved to: 28_May_202005_23PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 856.106934
66 2020-04-22 sub_region ... 66 898.349731
67 2020-04-23 sub_region ... 67 857.475891
68 2020-04-24 sub_region ... 68 861.468872
69 2020-04-25 sub_region ... 69 885.428162
70 2020-04-26 sub_region ... 70 897.812500
71 2020-04-27 sub_region ... 71 894.293274
72 2020-04-28 sub_region ... 72 929.067383
73 2020-04-29 sub_region ... 73 878.263550
74 2020-04-30 sub_region ... 74 869.549072
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/media/plotly/test_plot_20_314de4b9.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: dm02dl9g
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172341-dm02dl9g/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: dm02dl9g INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: b38vkw2p with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 1 use_mask: False wandb: Agent Started Run: b38vkw2p
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/b38vkw2p INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmIzOHZrdzJwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
15.36638443195261
The number of items in train is:
28
The loss for epoch 0
0.5487994439983075
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/media/graph/graph_0_summary_933ff711.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/media/graph
The running loss is: 19.380546062253416 The number of items in train is: 28 The loss for epoch 1 0.6921623593661934 The running loss is: 13.36081693135202 The number of items in train is: 28 The loss for epoch 2 0.4771720332625721
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-summary.json
The running loss is: 11.953934513032436 The number of items in train is: 28 The loss for epoch 3 0.4269262326083013 1 The running loss is: 12.163511937658768 The number of items in train is: 28 The loss for epoch 4 0.43441114063067027 The running loss is: 9.378338906506542 The number of items in train is: 28 The loss for epoch 5 0.33494067523237653 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-summary.json
The running loss is: 12.704894061491359 The number of items in train is: 28 The loss for epoch 6 0.4537462164818343 The running loss is: 11.437989893951453 The number of items in train is: 28 The loss for epoch 7 0.4084996390696948 1 The running loss is: 11.367679933318868 The number of items in train is: 28 The loss for epoch 8 0.40598856904710245
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-summary.json
The running loss is: 11.42585400538519 The number of items in train is: 28 The loss for epoch 9 0.40806621447804253 1 Data saved to: 28_May_202005_24PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['b38vkw2p']
Data saved to: 28_May_202005_24PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 802.753906
66 2020-04-22 sub_region ... 66 877.084839
67 2020-04-23 sub_region ... 67 830.681274
68 2020-04-24 sub_region ... 68 816.030273
69 2020-04-25 sub_region ... 69 821.921509
70 2020-04-26 sub_region ... 70 826.182373
71 2020-04-27 sub_region ... 71 846.830322
72 2020-04-28 sub_region ... 72 856.710266
73 2020-04-29 sub_region ... 73 857.412720
74 2020-04-30 sub_region ... 74 841.012085
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/media/plotly/test_plot_20_e30096eb.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: b38vkw2p
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172356-b38vkw2p/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: b38vkw2p INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: aj14ok4m with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 2 use_mask: True wandb: Agent Started Run: aj14ok4m
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/aj14ok4m INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmFqMTRvazRtOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
10.473457891494036
The number of items in train is:
27
The loss for epoch 0
0.38790584783311244
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/media/graph/graph_0_summary_321f1f13.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/media/graph
The running loss is: 12.50141921825707 The number of items in train is: 27 The loss for epoch 1 0.4630155266021137 1 The running loss is: 12.783501834142953 The number of items in train is: 27 The loss for epoch 2 0.47346303089418346
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-summary.json
The running loss is: 5.422599470242858 The number of items in train is: 27 The loss for epoch 3 0.20083701741640214 1 The running loss is: 5.399664411786944 The number of items in train is: 27 The loss for epoch 4 0.19998757080692384 The running loss is: 4.413807349279523 The number of items in train is: 27 The loss for epoch 5 0.16347434626961196
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-summary.json
The running loss is: 4.564219214487821 The number of items in train is: 27 The loss for epoch 6 0.16904515609214152 The running loss is: 4.552914118394256 The number of items in train is: 27 The loss for epoch 7 0.16862644882941688 1 The running loss is: 4.2418790506199 The number of items in train is: 27 The loss for epoch 8 0.15710663150444074
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-summary.json
2 The running loss is: 4.331287162844092 The number of items in train is: 27 The loss for epoch 9 0.16041804306829968 Data saved to: 28_May_202005_24PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['aj14ok4m']
Data saved to: 28_May_202005_24PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 839.469238
66 2020-04-22 sub_region ... 66 751.845520
67 2020-04-23 sub_region ... 67 796.254761
68 2020-04-24 sub_region ... 68 840.047058
69 2020-04-25 sub_region ... 69 856.757812
70 2020-04-26 sub_region ... 70 857.516724
71 2020-04-27 sub_region ... 71 805.222778
72 2020-04-28 sub_region ... 72 846.005188
73 2020-04-29 sub_region ... 73 746.322815
74 2020-04-30 sub_region ... 74 799.226501
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/media/plotly/test_plot_20_908a465b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: aj14ok4m
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172412-aj14ok4m/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: aj14ok4m INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ymutcajd with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 2 use_mask: False wandb: Agent Started Run: ymutcajd
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ymutcajd INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnltdXRjYWpkOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
10.56319503299892
The number of items in train is:
27
The loss for epoch 0
0.39122944566662665
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/media/graph/graph_0_summary_5dee85ff.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/media/graph
The running loss is: 12.365171766839921 The number of items in train is: 27 The loss for epoch 1 0.45796932469777485 1 The running loss is: 10.063402196858078 The number of items in train is: 27 The loss for epoch 2 0.37271859988363254 The running loss is: 5.191733076237142
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-summary.json
The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-history.jsonl
27 The loss for epoch 3 0.19228641023100526 1 The running loss is: 5.048446481116116 The number of items in train is: 27 The loss for epoch 4 0.1869794993005969 The running loss is: 4.6334264650940895 The number of items in train is: 27 The loss for epoch 5 0.17160838759607738 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-history.jsonl
The running loss is: 4.809368921909481 The number of items in train is: 27 The loss for epoch 6 0.17812477488553635 The running loss is: 4.469904407858849 The number of items in train is: 27 The loss for epoch 7 0.16555201510588327 1 The running loss is: 4.4312631585635245 The number of items in train is: 27 The loss for epoch 8 0.164120857724575
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-history.jsonl
The running loss is: 4.365964470198378 The number of items in train is: 27 The loss for epoch 9 0.1617023877851251 1 Data saved to: 28_May_202005_24PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_24PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['ymutcajd'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 820.661011
66 2020-04-22 sub_region ... 66 731.493591
67 2020-04-23 sub_region ... 67 777.978394
68 2020-04-24 sub_region ... 68 817.230530
69 2020-04-25 sub_region ... 69 833.734131
70 2020-04-26 sub_region ... 70 830.960205
71 2020-04-27 sub_region ... 71 781.981934
72 2020-04-28 sub_region ... 72 821.894836
73 2020-04-29 sub_region ... 73 732.720215
74 2020-04-30 sub_region ... 74 782.010132
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/media/plotly/test_plot_20_a6760719.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ymutcajd
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172428-ymutcajd/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ymutcajd INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: hzw7h1qu with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 3 use_mask: True wandb: Agent Started Run: hzw7h1qu
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/hzw7h1qu INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmh6dzdoMXF1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
22.41447990387678
The number of items in train is:
27
The loss for epoch 0
0.8301659223658068
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/media/graph/graph_0_summary_6d156013.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/media
The running loss is: 20.35543255507946 The number of items in train is: 27 The loss for epoch 1 0.7539049094473874 1 The running loss is: 19.27651271224022 The number of items in train is: 27 The loss for epoch 2 0.7139449152681563
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-history.jsonl
The running loss is: 12.351722501218319 The number of items in train is: 27 The loss for epoch 3 0.4574712037488266 1 The running loss is: 9.813126027584076 The number of items in train is: 27 The loss for epoch 4 0.36344911213274356 The running loss is: 6.200845863670111 The number of items in train is: 27 The loss for epoch 5 0.2296609579137078 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-history.jsonl
The running loss is: 5.99026171118021 The number of items in train is: 27 The loss for epoch 6 0.2218615448585263 The running loss is: 5.51112406142056 The number of items in train is: 27 The loss for epoch 7 0.20411570597853926 1 The running loss is: 5.962472440674901 The number of items in train is: 27 The loss for epoch 8 0.22083231261758893
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-history.jsonl
The running loss is: 5.207641955465078 The number of items in train is: 27 The loss for epoch 9 0.1928756279801881 1 Data saved to: 28_May_202005_24PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_24PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['hzw7h1qu'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 731.740784
66 2020-04-22 sub_region ... 66 738.606201
67 2020-04-23 sub_region ... 67 737.777405
68 2020-04-24 sub_region ... 68 726.210022
69 2020-04-25 sub_region ... 69 719.908691
70 2020-04-26 sub_region ... 70 719.150635
71 2020-04-27 sub_region ... 71 720.189941
72 2020-04-28 sub_region ... 72 731.834473
73 2020-04-29 sub_region ... 73 735.288940
74 2020-04-30 sub_region ... 74 731.156189
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/media/plotly/test_plot_20_3857be08.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: hzw7h1qu
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172444-hzw7h1qu/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: hzw7h1qu INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 36ic7346 with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 3 use_mask: False wandb: Agent Started Run: 36ic7346
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/36ic7346 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjM2aWM3MzQ2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
22.289740599691868
The number of items in train is:
27
The loss for epoch 0
0.8255459481367359
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/media/graph/graph_0_summary_3b05caf2.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/media/graph
The running loss is: 20.721937209367752 The number of items in train is: 27 The loss for epoch 1 0.7674791559025094 1 The running loss is: 19.501514554023743 The number of items in train is: 27 The loss for epoch 2 0.7222783168156942
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-summary.json
The running loss is: 13.022847175598145 The number of items in train is: 27 The loss for epoch 3 0.4823276731703017 1 The running loss is: 9.215275160968304 The number of items in train is: 27 The loss for epoch 4 0.3413064874432705 2 The running loss is: 6.453644126653671 The number of items in train is: 27 The loss for epoch 5 0.23902385654272856
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-summary.json
The running loss is: 6.3195782247930765 The number of items in train is: 27 The loss for epoch 6 0.23405845277011395 1 The running loss is: 5.538634759373963 The number of items in train is: 27 The loss for epoch 7 0.20513462071755417 The running loss is: 6.0211376789957285 The number of items in train is: 27 The loss for epoch 8 0.22300509922206402 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-summary.json
The running loss is: 5.4065384501591325 The number of items in train is: 27 The loss for epoch 9 0.2002421648207086 Data saved to: 28_May_202005_25PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_25PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['36ic7346']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 723.085754
66 2020-04-22 sub_region ... 66 731.862305
67 2020-04-23 sub_region ... 67 732.741089
68 2020-04-24 sub_region ... 68 723.526733
69 2020-04-25 sub_region ... 69 715.763306
70 2020-04-26 sub_region ... 70 713.154541
71 2020-04-27 sub_region ... 71 712.077576
72 2020-04-28 sub_region ... 72 725.158813
73 2020-04-29 sub_region ... 73 729.568176
74 2020-04-30 sub_region ... 74 726.045166
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/media/plotly/test_plot_20_967a32da.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 36ic7346
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172500-36ic7346/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 36ic7346 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: yndbm4kr with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 4 use_mask: True wandb: Agent Started Run: yndbm4kr
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/yndbm4kr INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnluZGJtNGtyOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
17.69617347419262
The number of items in train is:
26
The loss for epoch 0
0.6806220566997161
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/media/graph/graph_0_summary_a573ac99.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/media
The running loss is: 12.942448504269123 The number of items in train is: 26 The loss for epoch 1 0.4977864809334278 1 The running loss is: 11.007499657571316 The number of items in train is: 26 The loss for epoch 2 0.4233653714450506 2 The running loss is: 7.731042757630348 The number of items in train is: 26
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-history.jsonl
The loss for epoch 3 0.297347798370398 The running loss is: 4.677496463060379 The number of items in train is: 26 The loss for epoch 4 0.1799037101177069 1 The running loss is: 3.537908681668341 The number of items in train is: 26 The loss for epoch 5 0.13607341083339775 The running loss is: 3.6070269970223308 The number of items in train is: 26 The loss for epoch 6 0.13873180757778195
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-history.jsonl
1 The running loss is: 5.469477290287614 The number of items in train is: 26 The loss for epoch 7 0.21036451116490823 2 The running loss is: 4.373710255138576 The number of items in train is: 26 The loss for epoch 8 0.16821962519763753
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-history.jsonl
The running loss is: 4.385507521219552 The number of items in train is: 26 The loss for epoch 9 0.168673366200752 Data saved to: 28_May_202005_25PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_25PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['yndbm4kr'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 641.062012
66 2020-04-22 sub_region ... 66 649.469788
67 2020-04-23 sub_region ... 67 662.655884
68 2020-04-24 sub_region ... 68 663.000610
69 2020-04-25 sub_region ... 69 652.813110
70 2020-04-26 sub_region ... 70 647.735291
71 2020-04-27 sub_region ... 71 634.131409
72 2020-04-28 sub_region ... 72 641.305664
73 2020-04-29 sub_region ... 73 641.875549
74 2020-04-30 sub_region ... 74 656.309570
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/media/plotly/test_plot_20_7320596b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: yndbm4kr
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172511-yndbm4kr/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: yndbm4kr INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: y2e0r61g with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 4 use_mask: False wandb: Agent Started Run: y2e0r61g
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/y2e0r61g INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnkyZTByNjFnOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
17.936274334788322
The number of items in train is:
26
The loss for epoch 0
0.6898567051841662
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/media/graph/graph_0_summary_e6593ad8.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/media/graph
The running loss is: 12.563346169888973 The number of items in train is: 26 The loss for epoch 1 0.48320562191880667 1 The running loss is: 10.63600341975689 The number of items in train is: 26 The loss for epoch 2 0.4090770546060342 2 The running loss is: 7.032686905935407 The number of items in train is: 26 The loss for epoch 3 0.27048795792059255
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-summary.json
The running loss is: 4.1291360929608345 The number of items in train is: 26 The loss for epoch 4 0.1588129266523398 1 The running loss is: 3.5215289955958724 The number of items in train is: 26 The loss for epoch 5 0.13544342290753356 The running loss is: 4.55407845415175
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-history.jsonl
The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-summary.json
26 The loss for epoch 6 0.17515686362122113 1 The running loss is: 3.83646054379642 The number of items in train is: 26 The loss for epoch 7 0.14755617476140076 2 The running loss is: 4.684140918310732 The number of items in train is: 26 The loss for epoch 8 0.18015926608887428
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-summary.json
The running loss is: 7.284909620881081 The number of items in train is: 26 The loss for epoch 9 0.28018883157234925 1 Data saved to: 28_May_202005_25PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_25PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['y2e0r61g'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 811.324219
66 2020-04-22 sub_region ... 66 801.534241
67 2020-04-23 sub_region ... 67 837.117920
68 2020-04-24 sub_region ... 68 854.320984
69 2020-04-25 sub_region ... 69 834.952698
70 2020-04-26 sub_region ... 70 831.260376
71 2020-04-27 sub_region ... 71 782.874146
72 2020-04-28 sub_region ... 72 832.120972
73 2020-04-29 sub_region ... 73 785.114258
74 2020-04-30 sub_region ... 74 817.378418
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/media/plotly/test_plot_20_d2368e1b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: y2e0r61g
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172527-y2e0r61g/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: y2e0r61g INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: gzism3qq with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 5 use_mask: True wandb: Agent Started Run: gzism3qq
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/gzism3qq INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmd6aXNtM3FxOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
16.614094384014606
The number of items in train is:
26
The loss for epoch 0
0.639003630154408
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/media/graph/graph_0_summary_9dd2c671.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/media
The running loss is: 12.784377932548523 The number of items in train is: 26 The loss for epoch 1 0.49170684355955857 1 The running loss is: 10.317914854735136 The number of items in train is: 26 The loss for epoch 2 0.3968428790282745
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-history.jsonl
The running loss is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-summary.json
8.098226331174374 The number of items in train is: 26 The loss for epoch 3 0.31147024350670666 1 The running loss is: 5.721522863255814 The number of items in train is: 26 The loss for epoch 4 0.22005857166368514 The running loss is: 4.689897304400802 The number of items in train is: 26 The loss for epoch 5 0.180380665553877 The running loss is: 6.495792439207435
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-summary.json
The number of items in train is: 26 The loss for epoch 6 0.24983817073874748 1 The running loss is: 5.418599771335721 The number of items in train is: 26 The loss for epoch 7 0.20840768351291233 The running loss is: 4.195066409185529 The number of items in train is: 26 The loss for epoch 8 0.16134870804559726 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-summary.json
The running loss is: 5.375888114795089 The number of items in train is: 26 The loss for epoch 9 0.2067649274921188 2 Data saved to: 28_May_202005_25PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_25PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['gzism3qq'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 730.216797
66 2020-04-22 sub_region ... 66 741.966797
67 2020-04-23 sub_region ... 67 738.005005
68 2020-04-24 sub_region ... 68 724.475220
69 2020-04-25 sub_region ... 69 719.175964
70 2020-04-26 sub_region ... 70 717.805908
71 2020-04-27 sub_region ... 71 714.661133
72 2020-04-28 sub_region ... 72 731.062378
73 2020-04-29 sub_region ... 73 725.214966
74 2020-04-30 sub_region ... 74 723.802612
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/media/plotly/test_plot_20_54bf2aea.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: gzism3qq
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172543-gzism3qq/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: gzism3qq INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: lvl9ftc6 with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 5 use_mask: False wandb: Agent Started Run: lvl9ftc6
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/lvl9ftc6 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmx2bDlmdGM2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
16.377284348011017
The number of items in train is:
26
The loss for epoch 0
0.6298955518465775
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/media/graph/graph_0_summary_5ef9dede.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/media
The running loss is: 14.166790530085564 The number of items in train is: 26 The loss for epoch 1 0.5448765588494447 The running loss is: 9.428598899394274 The number of items in train is: 26 The loss for epoch 2 0.3626384192074721 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/wandb-history.jsonl
The running loss is: 6.6330753564834595 The number of items in train is: 26 The loss for epoch 3 0.2551182829416715 2 The running loss is: 5.67338194232434 The number of items in train is: 26 The loss for epoch 4 0.2182069977817054 3 Stopping model now Data saved to: 28_May_202005_26PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_26PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 339.937469
66 2020-04-22 sub_region ... 66 348.662567
67 2020-04-23 sub_region ... 67 321.414612
68 2020-04-24 sub_region ... 68 302.783997
69 2020-04-25 sub_region ... 69 324.099854
70 2020-04-26 sub_region ... 70 317.279297
71 2020-04-27 sub_region ... 71 371.108948
72 2020-04-28 sub_region ... 72 297.625122
73 2020-04-29 sub_region ... 73 384.647125
74 2020-04-30 sub_region ... 74 332.773010
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.wandb_agent:Running runs: ['lvl9ftc6']
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/media/plotly/test_plot_10_807ad6e3.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: lvl9ftc6
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/media/plotly/test_plot_all_11_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172559-lvl9ftc6/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: lvl9ftc6 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: acomglw4 with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 1 use_mask: True wandb: Agent Started Run: acomglw4
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/acomglw4 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmFjb21nbHc0OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/media/graph/graph_0_summary_0047ac6a.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/media
The running loss is: 16.114518700167537 The number of items in train is: 28 The loss for epoch 0 0.5755185250059834 The running loss is: 40.93923968449235 The number of items in train is: 28 The loss for epoch 1 1.462115703017584
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-summary.json
The running loss is: 17.72650630073622 The number of items in train is: 28 The loss for epoch 2 0.6330895107405793 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-summary.json
The running loss is: 13.82603452168405 The number of items in train is: 28 The loss for epoch 3 0.4937869472030018 The running loss is: 12.235369089990854 The number of items in train is: 28 The loss for epoch 4 0.43697746749967337
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-summary.json
The running loss is: 12.42129836557433 The number of items in train is: 28 The loss for epoch 5 0.4436177987705118 1
INFO:wandb.wandb_agent:Running runs: ['acomglw4']
The running loss is: 11.593332045944408 The number of items in train is: 28 The loss for epoch 6 0.4140475730694431
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-summary.json
The running loss is: 12.022986069088802 The number of items in train is: 28 The loss for epoch 7 0.42939235961031436 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-summary.json
The running loss is: 11.050711158488411 The number of items in train is: 28 The loss for epoch 8 0.3946682556603004 2 The running loss is: 10.95676115965398 The number of items in train is: 28 The loss for epoch 9 0.3913128985590707 3 Stopping model now Data saved to: 28_May_202005_26PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-summary.json
Data saved to: 28_May_202005_26PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 837.403687
66 2020-04-22 sub_region ... 66 830.898865
67 2020-04-23 sub_region ... 67 824.932678
68 2020-04-24 sub_region ... 68 836.498291
69 2020-04-25 sub_region ... 69 841.817627
70 2020-04-26 sub_region ... 70 846.289368
71 2020-04-27 sub_region ... 71 830.351929
72 2020-04-28 sub_region ... 72 852.223999
73 2020-04-29 sub_region ... 73 823.185059
74 2020-04-30 sub_region ... 74 832.774414
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/media/plotly/test_plot_20_754d0a08.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/media/plotly
wandb: Agent Finished Run: acomglw4
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172610-acomglw4/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: acomglw4 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ir0x5ngg with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 1 use_mask: False wandb: Agent Started Run: ir0x5ngg
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ir0x5ngg INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmlyMHg1bmdnOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/media/graph/graph_0_summary_b50c2bff.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/media/graph
The running loss is: 15.39689282909967 The number of items in train is: 28 The loss for epoch 0 0.5498890296107025 The running loss is: 44.767726235091686 The number of items in train is: 28 The loss for epoch 1 1.5988473655389888
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/wandb-history.jsonl
The running loss is: 22.245736518874764 The number of items in train is: 28 The loss for epoch 2 0.794490589959813 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/wandb-history.jsonl
The running loss is: 15.605747589841485 The number of items in train is: 28 The loss for epoch 3 0.5573481282086244 2 The running loss is: 18.130321642383933 The number of items in train is: 28 The loss for epoch 4 0.6475114872279976 3 Stopping model now Data saved to: 28_May_202005_26PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/wandb-history.jsonl
Data saved to: 28_May_202005_26PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['ir0x5ngg']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 410.513519
66 2020-04-22 sub_region ... 66 402.391907
67 2020-04-23 sub_region ... 67 409.347412
68 2020-04-24 sub_region ... 68 416.087708
69 2020-04-25 sub_region ... 69 416.266327
70 2020-04-26 sub_region ... 70 415.997223
71 2020-04-27 sub_region ... 71 405.508575
72 2020-04-28 sub_region ... 72 415.981689
73 2020-04-29 sub_region ... 73 393.018555
74 2020-04-30 sub_region ... 74 405.671448
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ir0x5ngg
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/media/plotly/test_plot_10_d4fb56b9.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/media/plotly/test_plot_all_11_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172625-ir0x5ngg/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ir0x5ngg INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: fjt2fktl with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 2 use_mask: True wandb: Agent Started Run: fjt2fktl
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/fjt2fktl INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmZqdDJma3RsOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/media/graph/graph_0_summary_03895c49.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/media/graph
The running loss is: 10.718430198729038 The number of items in train is: 27 The loss for epoch 0 0.3969788962492236 The running loss is: 28.698574017733335 The number of items in train is: 27 The loss for epoch 1 1.0629101488049384
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-history.jsonl
The running loss is: 16.018407717347145 The number of items in train is: 27 The loss for epoch 2 0.5932743599017462 1 The running loss is: 10.534781254827976 The number of items in train is: 27
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-summary.json
The loss for epoch 3
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-history.jsonl
0.39017708351214725 2 The running loss is: 5.385364955291152 The number of items in train is: 27 The loss for epoch 4 0.1994579613070797
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-history.jsonl
The running loss is: 7.625352939590812 The number of items in train is: 27 The loss for epoch 5 0.28242047924410413 1 The running loss is: 6.7238668041536584 The number of items in train is:
INFO:wandb.wandb_agent:Running runs: ['fjt2fktl']
27 The loss for epoch 6 0.2490321038575429
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-history.jsonl
The running loss is: 7.8772421116009355 The number of items in train is: 27 The loss for epoch 7 0.29174970783707166 The running loss is: 7.317102946341038
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-summary.json
The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-history.jsonl
27 The loss for epoch 8 0.27100381282744584 1 The running loss is: 5.420254968106747 The number of items in train is: 27 The loss for epoch 9 0.20075018400395359 Data saved to: 28_May_202005_26PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-history.jsonl
Data saved to: 28_May_202005_26PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 779.043823
66 2020-04-22 sub_region ... 66 778.755005
67 2020-04-23 sub_region ... 67 778.881592
68 2020-04-24 sub_region ... 68 779.025024
69 2020-04-25 sub_region ... 69 779.112366
70 2020-04-26 sub_region ... 70 779.083008
71 2020-04-27 sub_region ... 71 778.987671
72 2020-04-28 sub_region ... 72 778.948608
73 2020-04-29 sub_region ... 73 778.723877
74 2020-04-30 sub_region ... 74 778.891602
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: fjt2fktl
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/media/plotly/test_plot_20_03adc1c1.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172636-fjt2fktl/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: fjt2fktl INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: jt4p6a4c with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 2 use_mask: False wandb: Agent Started Run: jt4p6a4c
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/jt4p6a4c INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmp0NHA2YTRjOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/media/graph/graph_0_summary_b3c42c99.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/media/graph
The running loss is: 10.668751269578934 The number of items in train is: 27 The loss for epoch 0 0.3951389359103309 The running loss is: 28.521158926188946 The number of items in train is: 27 The loss for epoch 1 1.0563392194884795
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-history.jsonl
The running loss is: 14.176970317959785 The number of items in train is: 27 The loss for epoch 2 0.5250729747392513
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-history.jsonl
The running loss is: 6.591888493858278 The number of items in train is: 27 The loss for epoch 3 0.24414401829104732 1 The running loss is: 7.487370473332703 The number of items in train is: 27 The loss for epoch 4 0.27731001753084084
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-history.jsonl
The running loss is: 5.26125293970108 The number of items in train is: 27 The loss for epoch 5 0.1948612199889289
INFO:wandb.wandb_agent:Running runs: ['jt4p6a4c']
The running loss is: 6.058157247491181 The number of items in train is: 27 The loss for epoch 6 0.22437619435152523 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-history.jsonl
The running loss is: 4.886977478861809 The number of items in train is: 27 The loss for epoch 7 0.18099916588377069
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-history.jsonl
The running loss is: 8.049873136915267 The number of items in train is: 27 The loss for epoch 8 0.29814344951538024 1 The running loss is: 6.706313902512193 The number of items in train is: 27 The loss for epoch 9 0.24838199638934047 2 Data saved to: 28_May_202005_26PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-history.jsonl
Data saved to: 28_May_202005_26PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 755.005249
66 2020-04-22 sub_region ... 66 751.443115
67 2020-04-23 sub_region ... 67 753.844849
68 2020-04-24 sub_region ... 68 755.624390
69 2020-04-25 sub_region ... 69 755.841370
70 2020-04-26 sub_region ... 70 755.883606
71 2020-04-27 sub_region ... 71 752.995850
72 2020-04-28 sub_region ... 72 755.255066
73 2020-04-29 sub_region ... 73 751.328247
74 2020-04-30 sub_region ... 74 753.881714
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/media/plotly/test_plot_20_9c0dfd36.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/media/plotly
wandb: Agent Finished Run: jt4p6a4c
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172652-jt4p6a4c/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: jt4p6a4c INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 2gjy2l2o with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 3 use_mask: True wandb: Agent Started Run: 2gjy2l2o
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/2gjy2l2o INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjJnankybDJvOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/media/graph/graph_0_summary_53cb4522.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/media/graph
The running loss is: 23.849574618041515 The number of items in train is: 27 The loss for epoch 0 0.883317578445982 The running loss is: 20.568836465477943 The number of items in train is: 27 The loss for epoch 1 0.7618087579806646
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-summary.json
The running loss is: 19.27038937062025 The number of items in train is: 27 The loss for epoch 2 0.7137181248377871
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-summary.json
The running loss is: 17.25364562869072 The number of items in train is: 27 The loss for epoch 3 0.6390239121737303 1 The running loss is: 14.78194186091423 The number of items in train is: 27 The loss for epoch 4 0.5474793281820085
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-summary.json
The running loss is: 10.99204146116972 The number of items in train is: 27 The loss for epoch 5 0.40711264670998965 1
INFO:wandb.wandb_agent:Running runs: ['2gjy2l2o']
The running loss is: 34.4778664149344 The number of items in train is: 27 The loss for epoch 6 1.2769580153679405
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-summary.json
The running loss is: 18.340780273079872 The number of items in train is: 27 The loss for epoch 7 0.6792881582622174 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-summary.json
The running loss is: 18.76011623442173 The number of items in train is: 27 The loss for epoch 8 0.6948191197933974 2 The running loss is: 18.743604812771082 The number of items in train is: 27 The loss for epoch 9 0.6942075856581882 Data saved to: 28_May_202005_27PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_27PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 351.287323
66 2020-04-22 sub_region ... 66 351.352722
67 2020-04-23 sub_region ... 67 351.261353
68 2020-04-24 sub_region ... 68 351.227020
69 2020-04-25 sub_region ... 69 351.256317
70 2020-04-26 sub_region ... 70 351.258667
71 2020-04-27 sub_region ... 71 351.388000
72 2020-04-28 sub_region ... 72 351.195374
73 2020-04-29 sub_region ... 73 351.413818
74 2020-04-30 sub_region ... 74 351.286987
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/media/plotly/test_plot_20_3234c249.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 2gjy2l2o
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172708-2gjy2l2o/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 2gjy2l2o INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ei4en7rs with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 3 use_mask: False wandb: Agent Started Run: ei4en7rs
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ei4en7rs INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmVpNGVuN3JzOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/media/graph/graph_0_summary_4e0b46a5.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/media/graph
The running loss is: 23.785254307091236 The number of items in train is: 27 The loss for epoch 0 0.8809353447070828 The running loss is: 20.365187123417854 The number of items in train is: 27 The loss for epoch 1 0.7542661897562168
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-summary.json
The running loss is: 19.407820589840412 The number of items in train is: 27 The loss for epoch 2 0.7188081699940894
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-summary.json
The running loss is: 16.047366067767143 The number of items in train is: 27 The loss for epoch 3 0.594346891398783 1 The running loss is: 16.049381844699383 The number of items in train is: 27 The loss for epoch 4 0.5944215498036809
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-summary.json
The running loss is: 8.695411691442132 The number of items in train is: 27 The loss for epoch 5 0.3220522848682271 1
INFO:wandb.wandb_agent:Running runs: ['ei4en7rs']
The running loss is: 11.989098764955997 The number of items in train is: 27 The loss for epoch 6 0.4440406949983703
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-summary.json
The running loss is: 12.975145380944014 The number of items in train is: 27 The loss for epoch 7 0.48056094003496347 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-summary.json
The running loss is: 6.981494817882776 The number of items in train is: 27 The loss for epoch 8 0.25857388214380655 The running loss is: 7.491051498800516 The number of items in train is: 27 The loss for epoch 9 0.27744635180742655 1 Data saved to: 28_May_202005_27PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-summary.json
Data saved to: 28_May_202005_27PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 709.696960
66 2020-04-22 sub_region ... 66 709.161499
67 2020-04-23 sub_region ... 67 708.643188
68 2020-04-24 sub_region ... 68 708.597656
69 2020-04-25 sub_region ... 69 708.908020
70 2020-04-26 sub_region ... 70 708.755493
71 2020-04-27 sub_region ... 71 709.836182
72 2020-04-28 sub_region ... 72 708.476990
73 2020-04-29 sub_region ... 73 709.191528
74 2020-04-30 sub_region ... 74 708.709473
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ei4en7rs
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/media/plotly/test_plot_20_509b02c0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172726-ei4en7rs/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ei4en7rs INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 2ahqajh0 with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 4 use_mask: True wandb: Agent Started Run: 2ahqajh0
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/2ahqajh0 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjJhaHFhamgwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/media/graph/graph_0_summary_28353f47.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/media/graph
The running loss is:
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/media
25.31017677485943 The number of items in train is: 26 The loss for epoch 0 0.9734683374945934 The running loss is: 17.691114902496338 The number of items in train is: 26 The loss for epoch 1 0.6804274962498591 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-summary.json
The running loss is: 21.284134477376938 The number of items in train is: 26 The loss for epoch 2 0.8186205568221899 The running loss is: 14.811767756938934 The number of items in train is: 26 The loss for epoch 3 0.5696833752668821
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-summary.json
The running loss is: 12.015214286744595 The number of items in train is: 26 The loss for epoch 4 0.46212362641325366 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-summary.json
The running loss is: 10.353260762989521 The number of items in train is: 26 The loss for epoch 5 0.3982023370380585 2 The running loss is: 7.667264841496944 The number of items in train is: 26 The loss for epoch 6 0.2948948015960363 3 Stopping model now Data saved to: 28_May_202005_27PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['2ahqajh0'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-summary.json
Data saved to: 28_May_202005_27PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 318.850616
66 2020-04-22 sub_region ... 66 316.706177
67 2020-04-23 sub_region ... 67 320.013977
68 2020-04-24 sub_region ... 68 321.565765
69 2020-04-25 sub_region ... 69 321.352051
70 2020-04-26 sub_region ... 70 320.382446
71 2020-04-27 sub_region ... 71 318.602081
72 2020-04-28 sub_region ... 72 318.317017
73 2020-04-29 sub_region ... 73 318.346680
74 2020-04-30 sub_region ... 74 320.484741
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 2ahqajh0
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/media/plotly/test_plot_14_b1c54b64.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/media/plotly/test_plot_all_15_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172742-2ahqajh0/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 2ahqajh0 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 9cofme0n with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 4 use_mask: False wandb: Agent Started Run: 9cofme0n
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/9cofme0n INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjljb2ZtZTBuOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/media/graph/graph_0_summary_9235a771.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/media
The running loss is: 25.090075597167015 The number of items in train is: 26 The loss for epoch 0 0.9650029075833467 The running loss is: 17.45224541425705 The number of items in train is: 26 The loss for epoch 1 0.6712402082406558 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-summary.json
The running loss is: 21.34785506129265 The number of items in train is: 26 The loss for epoch 2 0.8210713485112557 The running loss is: 14.617094554007053 The number of items in train is: 26 The loss for epoch 3 0.5621959443848866
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-summary.json
1 The running loss is: 12.857044279575348 The number of items in train is: 26 The loss for epoch 4 0.4945017030605903
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-summary.json
The running loss is: 10.908861950039864 The number of items in train is: 26 The loss for epoch 5 0.4195716134630717 1
INFO:wandb.wandb_agent:Running runs: ['9cofme0n'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-summary.json
The running loss is: 7.671748608350754 The number of items in train is: 26 The loss for epoch 6 0.2950672541673367 The running loss is: 6.437005430459976 The number of items in train is: 26 The loss for epoch 7 0.2475771319407683
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-summary.json
The running loss is: 7.865247845649719 The number of items in train is: 26 The loss for epoch 8 0.3025095325249892 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-summary.json
The running loss is: 5.2042678743600845 The number of items in train is: 26 The loss for epoch 9 0.2001641490138494 Data saved to: 28_May_202005_28PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_28PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/config.yaml /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 570.512756
66 2020-04-22 sub_region ... 66 571.143372
67 2020-04-23 sub_region ... 67 570.903015
68 2020-04-24 sub_region ... 68 570.691162
69 2020-04-25 sub_region ... 69 570.595093
70 2020-04-26 sub_region ... 70 570.591003
71 2020-04-27 sub_region ... 71 570.901611
72 2020-04-28 sub_region ... 72 570.663696
73 2020-04-29 sub_region ... 73 571.259827
74 2020-04-30 sub_region ... 74 571.026001
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/media/plotly/test_plot_20_96e49c04.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 9cofme0n
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172758-9cofme0n/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 9cofme0n INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: dg2wr63w with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 5 use_mask: True wandb: Agent Started Run: dg2wr63w
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/dg2wr63w INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmRnMndyNjN3OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/media/graph/graph_0_summary_16b3fa38.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/media
The running loss is: 24.998789221048355 The number of items in train is: 26 The loss for epoch 0 0.9614918931172445 The running loss is: 19.62556444108486 The number of items in train is: 26 The loss for epoch 1 0.754829401580187 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-summary.json
The running loss is: 19.021592140197754 The number of items in train is: 26 The loss for epoch 2 0.7315996976999136 The running loss is: 13.53062754869461 The number of items in train is: 26
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-history.jsonl
The loss for epoch 3
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-summary.json
0.5204087518728696 The running loss is: 11.939093425869942 The number of items in train is: 26 The loss for epoch 4 0.45919590099499774
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-summary.json
The running loss is: 12.370361641049385 The number of items in train is: 26 The loss for epoch 5 0.475783140040361 1
INFO:wandb.wandb_agent:Running runs: ['dg2wr63w']
The running loss is: 10.105389013886452 The number of items in train is: 26 The loss for epoch 6 0.388668808226402
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-summary.json
The running loss is: 5.374523870646954 The number of items in train is: 26 The loss for epoch 7 0.20671245656334436 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-summary.json
The running loss is: 4.875691753812134 The number of items in train is: 26 The loss for epoch 8 0.18752660591585132 The running loss is: 7.123469600221142 The number of items in train is: 26 The loss for epoch 9 0.27397960000850546 1 Data saved to: 28_May_202005_28PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-summary.json
Data saved to: 28_May_202005_28PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 697.734924
66 2020-04-22 sub_region ... 66 699.246887
67 2020-04-23 sub_region ... 67 698.447754
68 2020-04-24 sub_region ... 68 697.592102
69 2020-04-25 sub_region ... 69 697.349731
70 2020-04-26 sub_region ... 70 697.215332
71 2020-04-27 sub_region ... 71 698.537048
72 2020-04-28 sub_region ... 72 697.177002
73 2020-04-29 sub_region ... 73 699.459961
74 2020-04-30 sub_region ... 74 698.878723
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/media/plotly/test_plot_20_70b23860.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: dg2wr63w
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172814-dg2wr63w/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: dg2wr63w INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: okra1ju6 with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 5 use_mask: False wandb: Agent Started Run: okra1ju6
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/okra1ju6 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm9rcmExanU2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/media/graph/graph_0_summary_50893349.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/media
The running loss is: 24.827137678861618 The number of items in train is: 26 The loss for epoch 0 0.9548899107254468 The running loss is: 19.719864428043365 The number of items in train is: 26 The loss for epoch 1 0.758456324155514 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-history.jsonl
The running loss is: 18.63474379479885 The number of items in train is: 26 The loss for epoch 2 0.7167209151845712 The running loss is: 13.806637480854988 The number of items in train is: 26 The loss for epoch 3 0.5310245184944227
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-history.jsonl
1 The running loss is: 11.331828974187374 The number of items in train is: 26 The loss for epoch 4 0.4358395759302836
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-history.jsonl
The running loss is: 10.775343157351017 The number of items in train is: 26 The loss for epoch 5 0.4144362752827314 1 The running loss is: 10.32153557986021 The number of items in train is: 26 The loss for epoch 6 0.3969821376869312 2
INFO:wandb.wandb_agent:Running runs: ['okra1ju6'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-history.jsonl
The running loss is: 7.037654765415937 The number of items in train is: 26 The loss for epoch 7 0.2706790294390745 The running loss is: 14.718151992186904 The number of items in train is: 26 The loss for epoch 8 0.5660827689302655 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-history.jsonl
The running loss is: 5.737017912324518 The number of items in train is: 26 The loss for epoch 9 0.22065453508940452 Data saved to: 28_May_202005_28PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_28PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 632.474121
66 2020-04-22 sub_region ... 66 634.329102
67 2020-04-23 sub_region ... 67 632.773071
68 2020-04-24 sub_region ... 68 631.713074
69 2020-04-25 sub_region ... 69 631.196228
70 2020-04-26 sub_region ... 70 631.151917
71 2020-04-27 sub_region ... 71 632.160889
72 2020-04-28 sub_region ... 72 631.541016
73 2020-04-29 sub_region ... 73 633.977478
74 2020-04-30 sub_region ... 74 632.605713
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/media/plotly/test_plot_20_a4b0535b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: okra1ju6
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172833-okra1ju6/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: okra1ju6 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: sgw7c1ep with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 1 use_mask: True wandb: Agent Started Run: sgw7c1ep
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/sgw7c1ep INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnNndzdjMWVwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/media/graph/graph_0_summary_3ce5022b.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/media
The running loss is: 16.056726850569248 The number of items in train is: 28 The loss for epoch 0 0.5734545303774732
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-summary.json
The running loss is: 46.208357490599155 The number of items in train is: 28 The loss for epoch 1 1.6502984818071127
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-summary.json
The running loss is: 29.032308392226696 The number of items in train is: 28 The loss for epoch 2 1.0368681568652391
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-summary.json
The running loss is: 28.841471418738365 The number of items in train is: 28 The loss for epoch 3 1.0300525506692273
INFO:wandb.wandb_agent:Running runs: ['sgw7c1ep'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-summary.json
The running loss is: 28.338371301069856 The number of items in train is: 28 The loss for epoch 4 1.0120846893239235 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-summary.json
The running loss is: 28.69572694809176 The number of items in train is: 28 The loss for epoch 5 1.0248473910032772 2 The running loss is: 19.646056853234768 The number of items in train is: 28 The loss for epoch 6 0.7016448876155275
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-history.jsonl
3
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-summary.json
Stopping model now Data saved to: 28_May_202005_28PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_28PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 429.154022
66 2020-04-22 sub_region ... 66 428.729492
67 2020-04-23 sub_region ... 67 429.150635
68 2020-04-24 sub_region ... 68 429.542450
69 2020-04-25 sub_region ... 69 429.516174
70 2020-04-26 sub_region ... 70 429.553772
71 2020-04-27 sub_region ... 71 428.790649
72 2020-04-28 sub_region ... 72 429.774963
73 2020-04-29 sub_region ... 73 428.176636
74 2020-04-30 sub_region ... 74 428.942596
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/media/plotly/test_plot_14_bdfadf79.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: sgw7c1ep
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/media/plotly/test_plot_all_15_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172848-sgw7c1ep/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: sgw7c1ep INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: xptik3ho with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 1 use_mask: False wandb: Agent Started Run: xptik3ho
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/xptik3ho INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnhwdGlrM2hvOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/media/graph/graph_0_summary_c719c3e7.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/media/graph
The running loss is: 13.879746047779918 The number of items in train is: 28 The loss for epoch 0 0.4957052159921399
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-summary.json
The running loss is: 50.16468261182308 The number of items in train is: 28 The loss for epoch 1 1.7915958075651102
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-summary.json
The running loss is: 28.752860818058252 The number of items in train is: 28 The loss for epoch 2 1.0268878863592232
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-summary.json
The running loss is: 28.75844579562545 The number of items in train is: 28 The loss for epoch 3 1.027087349843766
INFO:wandb.wandb_agent:Running runs: ['xptik3ho'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-summary.json
The running loss is: 28.228818399831653 The number of items in train is: 28 The loss for epoch 4 1.0081720857082732 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-summary.json
The running loss is: 28.275303000351414 The number of items in train is: 28 The loss for epoch 5 1.0098322500125505 2 The running loss is: 19.328091056086123 The number of items in train is: 28 The loss for epoch 6 0.6902889662887901 3 Stopping model now
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-summary.json
Data saved to: 28_May_202005_29PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_29PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 429.017181
66 2020-04-22 sub_region ... 66 428.986389
67 2020-04-23 sub_region ... 67 429.089111
68 2020-04-24 sub_region ... 68 429.224731
69 2020-04-25 sub_region ... 69 429.164246
70 2020-04-26 sub_region ... 70 429.191895
71 2020-04-27 sub_region ... 71 428.905151
72 2020-04-28 sub_region ... 72 429.181915
73 2020-04-29 sub_region ... 73 428.645477
74 2020-04-30 sub_region ... 74 428.955353
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/media/plotly/test_plot_14_562144b7.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/media/plotly
wandb: Agent Finished Run: xptik3ho
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/media/plotly/test_plot_all_15_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172904-xptik3ho/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: xptik3ho INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: rhkbw3nt with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 2 use_mask: True wandb: Agent Started Run: rhkbw3nt
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/rhkbw3nt INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnJoa2J3M250OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/media/graph/graph_0_summary_a092fa95.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/media/graph
The running loss is: 9.515796359628439 The number of items in train is: 27 The loss for epoch 0 0.3524369022084607
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-summary.json
The running loss is: 31.958917625248432 The number of items in train is: 27 The loss for epoch 1 1.1836636157499418
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-summary.json
The running loss is: 17.400996170938015 The number of items in train is: 27 The loss for epoch 2 0.6444813396643709 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-summary.json
The running loss is: 17.490230657160282 The number of items in train is: 27 The loss for epoch 3 0.647786320635566
INFO:wandb.wandb_agent:Running runs: ['rhkbw3nt'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-summary.json
The running loss is: 17.88597560673952 The number of items in train is: 27 The loss for epoch 4 0.6624435409903526
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-summary.json
The running loss is: 17.275350011885166 The number of items in train is: 27 The loss for epoch 5 0.6398277782179691 The running loss is: 17.80729231238365 The number of items in train is: 27 The loss for epoch 6 0.6595293449030982
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-summary.json
The running loss is: 17.339848961681128 The number of items in train is: 27 The loss for epoch 7 0.6422166282104121
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-summary.json
The running loss is: 16.749763071537018 The number of items in train is: 27 The loss for epoch 8 0.6203615952421118
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-summary.json
The running loss is: 17.160026656463742 The number of items in train is: 27 The loss for epoch 9 0.6355565428319905 Data saved to: 28_May_202005_29PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-summary.json
Data saved to: 28_May_202005_29PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 452.964355
66 2020-04-22 sub_region ... 66 448.373505
67 2020-04-23 sub_region ... 67 451.731903
68 2020-04-24 sub_region ... 68 454.205078
69 2020-04-25 sub_region ... 69 454.852173
70 2020-04-26 sub_region ... 70 454.945068
71 2020-04-27 sub_region ... 71 451.557831
72 2020-04-28 sub_region ... 72 453.470245
73 2020-04-29 sub_region ... 73 448.514465
74 2020-04-30 sub_region ... 74 451.503876
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/media/plotly/test_plot_20_da2b038c.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/media/plotly
wandb: Agent Finished Run: rhkbw3nt
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172920-rhkbw3nt/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: rhkbw3nt INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: vt6j1jur with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 2 use_mask: False wandb: Agent Started Run: vt6j1jur
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/vt6j1jur INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnZ0NmoxanVyOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/media/graph/graph_0_summary_d7fd0963.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/media
The running loss is: 9.540501032024622 The number of items in train is: 27 The loss for epoch 0 0.353351890074986
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-history.jsonl
The running loss is: 32.01947431452572 The number of items in train is: 27 The loss for epoch 1 1.1859064560935453
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-history.jsonl
The running loss is: 17.433912321925163 The number of items in train is: 27 The loss for epoch 2 0.6457004563675987 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-history.jsonl
The running loss is: 17.372653737664223 The number of items in train is: 27 The loss for epoch 3 0.6434316199134897
INFO:wandb.wandb_agent:Running runs: ['vt6j1jur']
The running loss is: 17.937954049557447 The number of items in train is: 27 The loss for epoch 4 0.6643686685021277
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-history.jsonl
The running loss is: 17.54516338557005 The number of items in train is: 27 The loss for epoch 5 0.649820866132224
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-history.jsonl
The running loss is: 17.718922704458237 The number of items in train is: 27 The loss for epoch 6 0.6562563964614162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-history.jsonl
The running loss is: 17.30660080164671 The number of items in train is: 27 The loss for epoch 7 0.6409852148758041
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-history.jsonl
The running loss is: 16.718926943838596 The number of items in train is: 27 The loss for epoch 8 0.6192195164384665
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-history.jsonl
The running loss is: 16.827449632808566 The number of items in train is: 27 The loss for epoch 9 0.6232388752892062 Data saved to: 28_May_202005_29PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_29PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 457.992615
66 2020-04-22 sub_region ... 66 454.040710
67 2020-04-23 sub_region ... 67 456.812286
68 2020-04-24 sub_region ... 68 459.140167
69 2020-04-25 sub_region ... 69 459.434723
70 2020-04-26 sub_region ... 70 459.551544
71 2020-04-27 sub_region ... 71 456.045074
72 2020-04-28 sub_region ... 72 458.655548
73 2020-04-29 sub_region ... 73 453.452576
74 2020-04-30 sub_region ... 74 456.783661
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: vt6j1jur
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/media/plotly/test_plot_20_c566c54f.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172941-vt6j1jur/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: vt6j1jur INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: j1akj0pq with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 3 use_mask: True wandb: Agent Started Run: j1akj0pq
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/j1akj0pq INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmoxYWtqMHBxOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/media/graph/graph_0_summary_2890e43d.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/media/graph
The running loss is: 31.62208504229784 The number of items in train is: 27 The loss for epoch 0 1.17118833489992
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-summary.json
The running loss is: 19.408665716648102 The number of items in train is: 27 The loss for epoch 1 0.7188394709869668 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-summary.json
The running loss is: 20.407186299562454 The number of items in train is: 27 The loss for epoch 2 0.7558217147986094
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-summary.json
The running loss is: 20.506580412387848 The number of items in train is: 27 The loss for epoch 3 0.759502978236587
INFO:wandb.wandb_agent:Running runs: ['j1akj0pq'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-history.jsonl
The running loss is: 20.30820331722498 The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-summary.json
27 The loss for epoch 4 0.7521556784157399 The running loss is: 19.309663198888302 The number of items in train is: 27 The loss for epoch 5 0.7151727110699371 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-summary.json
The running loss is: 18.341759957373142 The number of items in train is: 27 The loss for epoch 6 0.679324442865672
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-summary.json
The running loss is: 17.737579837441444 The number of items in train is: 27 The loss for epoch 7 0.6569474013867201 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-summary.json
The running loss is: 19.28192499279976 The number of items in train is: 27 The loss for epoch 8 0.7141453701036947
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-summary.json
The running loss is: 18.758581429719925 The number of items in train is: 27 The loss for epoch 9 0.694762275174812 1 Data saved to: 28_May_202005_30PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_30PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 337.341278
66 2020-04-22 sub_region ... 66 337.345581
67 2020-04-23 sub_region ... 67 337.355377
68 2020-04-24 sub_region ... 68 337.360901
69 2020-04-25 sub_region ... 69 337.360565
70 2020-04-26 sub_region ... 70 337.355988
71 2020-04-27 sub_region ... 71 337.354065
72 2020-04-28 sub_region ... 72 337.343140
73 2020-04-29 sub_region ... 73 337.349548
74 2020-04-30 sub_region ... 74 337.355652
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/media/plotly/test_plot_20_e91edfe1.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: j1akj0pq
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_172957-j1akj0pq/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: j1akj0pq INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 30ix45tz with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 3 use_mask: False wandb: Agent Started Run: 30ix45tz
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/30ix45tz INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjMwaXg0NXR6OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/media/graph/graph_0_summary_a11268f2.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/media/graph
The running loss is: 31.686953715980053 The number of items in train is: 27 The loss for epoch 0 1.1735908783696316
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-history.jsonl
The running loss is: 19.326208144426346 The number of items in train is: 27 The loss for epoch 1 0.7157854868306054 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-history.jsonl
The running loss is: 20.44809077680111 The number of items in train is: 27 The loss for epoch 2 0.7573366954370782
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-history.jsonl
The running loss is: 20.70396327972412 The number of items in train is: 27 The loss for epoch 3 0.7668134548045971
INFO:wandb.wandb_agent:Running runs: ['30ix45tz'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-history.jsonl
The running loss is: 20.114036343991756 The number of items in train is: 27 The loss for epoch 4 0.7449643090367317 1 The running loss is: 19.89861784130335 The number of items in train is: 27 The loss for epoch 5 0.7369858459741981
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-history.jsonl
The running loss is: 18.967432893812656 The number of items in train is: 27 The loss for epoch 6 0.7024975145856539 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-history.jsonl
The running loss is: 19.17745415121317 The number of items in train is: 27 The loss for epoch 7 0.7102760796745619
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-history.jsonl
The running loss is: 19.25604397803545 The number of items in train is: 27 The loss for epoch 8 0.713186814001313
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-history.jsonl
The running loss is: 18.80481818318367 The number of items in train is: 27 The loss for epoch 9 0.6964747475253211 Data saved to: 28_May_202005_30PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_30PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 353.958618
66 2020-04-22 sub_region ... 66 353.994934
67 2020-04-23 sub_region ... 67 353.966064
68 2020-04-24 sub_region ... 68 353.948608
69 2020-04-25 sub_region ... 69 353.950348
70 2020-04-26 sub_region ... 70 353.951172
71 2020-04-27 sub_region ... 71 353.985535
72 2020-04-28 sub_region ... 72 353.945618
73 2020-04-29 sub_region ... 73 354.007202
74 2020-04-30 sub_region ... 74 353.972107
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/media/plotly/test_plot_20_384740d4.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 30ix45tz
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173018-30ix45tz/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 30ix45tz INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 8r2cy0dd with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 4 use_mask: True wandb: Agent Started Run: 8r2cy0dd
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/8r2cy0dd INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjhyMmN5MGRkOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/media/graph/graph_0_summary_6452539b.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/media/graph
The running loss is: 28.677166178822517 The number of items in train is: 26 The loss for epoch 0 1.1029679299547122
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-history.jsonl
The running loss is: 19.185990750789642 The number of items in train is: 26 The loss for epoch 1 0.737922721184217
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-history.jsonl
The running loss is: 20.213715583086014 The number of items in train is: 26 The loss for epoch 2 0.777450599349462
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-history.jsonl
The running loss is: 15.579263865947723 The number of items in train is: 26 The loss for epoch 3 0.5992024563826047 1
INFO:wandb.wandb_agent:Running runs: ['8r2cy0dd']
The running loss is: 15.614615187048912 The number of items in train is: 26 The loss for epoch 4 0.6005621225788043
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-history.jsonl
The running loss is: 14.192182525992393 The number of items in train is: 26 The loss for epoch 5 0.5458531740766305 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-history.jsonl
The running loss is: 17.893233135342598 The number of items in train is: 26 The loss for epoch 6 0.6882012744362538
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-history.jsonl
The running loss is: 15.887526050209999 The number of items in train is: 26 The loss for epoch 7 0.6110586942388461
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-history.jsonl
The running loss is: 13.056910991668701 The number of items in train is: 26 The loss for epoch 8 0.50218888429495 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-history.jsonl
The running loss is: 10.211280450224876 The number of items in train is: 26 The loss for epoch 9 0.39274155577787984 2 Data saved to: 28_May_202005_30PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_30PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 409.355835
66 2020-04-22 sub_region ... 66 409.509674
67 2020-04-23 sub_region ... 67 409.471558
68 2020-04-24 sub_region ... 68 409.399261
69 2020-04-25 sub_region ... 69 409.354004
70 2020-04-26 sub_region ... 70 409.365173
71 2020-04-27 sub_region ... 71 409.424011
72 2020-04-28 sub_region ... 72 409.376404
73 2020-04-29 sub_region ... 73 409.552612
74 2020-04-30 sub_region ... 74 409.473022
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/media/plotly/test_plot_20_eefb9a0f.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 8r2cy0dd
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173035-8r2cy0dd/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 8r2cy0dd INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 6yrex3hf with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 4 use_mask: False wandb: Agent Started Run: 6yrex3hf
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/6yrex3hf INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjZ5cmV4M2hmOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/media/graph/graph_0_summary_13008330.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/media
The running loss is: 28.61015349626541 The number of items in train is: 26 The loss for epoch 0 1.1003905190871313
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-history.jsonl
The running loss is: 19.129196166992188 The number of items in train is: 26 The loss for epoch 1 0.7357383141150842
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-history.jsonl
The running loss is: 20.223093792796135 The number of items in train is: 26 The loss for epoch 2 0.7778112997229283
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-history.jsonl
The running loss is: 15.592137679457664 The number of items in train is: 26 The loss for epoch 3 0.599697603056064 1
INFO:wandb.wandb_agent:Running runs: ['6yrex3hf']
The running loss is: 14.741975575685501 The number of items in train is: 26 The loss for epoch 4 0.5669990606032885
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-history.jsonl
The running loss is: 14.141681730747223 The number of items in train is: 26 The loss for epoch 5 0.5439108357979701 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-history.jsonl
The running loss is: 17.65192748606205 The number of items in train is: 26 The loss for epoch 6 0.6789202879254634
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-history.jsonl
The running loss is: 13.934842199087143 The number of items in train is: 26 The loss for epoch 7 0.5359554691956594 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-history.jsonl
The running loss is: 10.86246071755886 The number of items in train is: 26 The loss for epoch 8 0.4177869506753408
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-history.jsonl
The running loss is: 9.102634258568287 The number of items in train is: 26 The loss for epoch 9 0.3501013176372418 1 Data saved to: 28_May_202005_31PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_31PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/config.yaml
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-history.jsonl
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 475.105896
66 2020-04-22 sub_region ... 66 475.061249
67 2020-04-23 sub_region ... 67 475.085632
68 2020-04-24 sub_region ... 68 475.112213
69 2020-04-25 sub_region ... 69 475.114807
70 2020-04-26 sub_region ... 70 475.139618
71 2020-04-27 sub_region ... 71 475.079346
72 2020-04-28 sub_region ... 72 475.173126
73 2020-04-29 sub_region ... 73 475.043427
74 2020-04-30 sub_region ... 74 475.085358
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/media/plotly/test_plot_20_114ca43c.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 6yrex3hf
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173056-6yrex3hf/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 6yrex3hf INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: zi373v86 with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 5 use_mask: True wandb: Agent Started Run: zi373v86
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/zi373v86 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnppMzczdjg2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/media/graph/graph_0_summary_cfdce5d6.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/media
The running loss is: 26.287698671221733 The number of items in train is: 26 The loss for epoch 0 1.0110653335085282
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-summary.json
The running loss is: 22.019511476159096 The number of items in train is: 26 The loss for epoch 1 0.8469042875445806
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-summary.json
The running loss is: 19.7527959048748 The number of items in train is: 26 The loss for epoch 2 0.7597229194182616
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-summary.json
The running loss is: 15.987582370638847 The number of items in train is: 26 The loss for epoch 3 0.6149070142553403 1
INFO:wandb.wandb_agent:Running runs: ['zi373v86']
The running loss is: 15.31729331612587 The number of items in train is: 26 The loss for epoch 4 0.5891266660048411
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-summary.json
The running loss is: 13.650041669607162 The number of items in train is: 26 The loss for epoch 5 0.5250016026771985 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-summary.json
The running loss is: 13.348268389701843 The number of items in train is: 26 The loss for epoch 6 0.5133949380654556 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-summary.json
The running loss is: 11.254892647266388 The number of items in train is: 26 The loss for epoch 7 0.4328804864333226 3 Stopping model now Data saved to: 28_May_202005_31PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_31PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 347.929901
66 2020-04-22 sub_region ... 66 348.376862
67 2020-04-23 sub_region ... 67 348.171692
68 2020-04-24 sub_region ... 68 347.880554
69 2020-04-25 sub_region ... 69 347.761993
70 2020-04-26 sub_region ... 70 347.774658
71 2020-04-27 sub_region ... 71 348.046265
72 2020-04-28 sub_region ... 72 347.813934
73 2020-04-29 sub_region ... 73 348.664154
74 2020-04-30 sub_region ... 74 348.247253
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/media/plotly/test_plot_16_3f181331.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: zi373v86
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/media/plotly/test_plot_all_17_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173117-zi373v86/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: zi373v86 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ock9vc16 with config: batch_size: 2 forecast_history: 10 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 5 use_mask: False wandb: Agent Started Run: ock9vc16
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ock9vc16 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm9jazl2YzE2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/media/graph/graph_0_summary_7dd8cccf.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/media
The running loss is: 26.332314282655716 The number of items in train is: 26 The loss for epoch 0 1.0127813185636814
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-history.jsonl
The running loss is: 21.822422578930855 The number of items in train is: 26 The loss for epoch 1 0.8393239453434944
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-history.jsonl
The running loss is: 19.969776809215546 The number of items in train is: 26 The loss for epoch 2 0.7680683388159826
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-history.jsonl
The running loss is: 15.921830900013447 The number of items in train is: 26 The loss for epoch 3 0.6123781115389787 1
INFO:wandb.wandb_agent:Running runs: ['ock9vc16']
The running loss is: 15.494909837841988 The number of items in train is: 26 The loss for epoch 4 0.5959580706862303
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-history.jsonl
The running loss is: 14.208502128720284 The number of items in train is: 26 The loss for epoch 5 0.5464808511046263 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-history.jsonl
The running loss is: 16.151983082294464 The number of items in train is: 26 The loss for epoch 6 0.621230118549787 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-history.jsonl
The running loss is: 12.41115165501833 The number of items in train is: 26 The loss for epoch 7 0.4773519867314742
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-history.jsonl
The running loss is: 9.447976488620043 The number of items in train is: 26 The loss for epoch 8 0.3633837111007709 The running loss is: 9.989659074693918 The number of items in train is: 26
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-summary.json
The loss for epoch 9 0.3842176567189969
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-history.jsonl
1 Data saved to: 28_May_202005_31PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_31PM_model.pth interpolate should be below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-summary.json
Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-history.jsonl
CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 578.508301
66 2020-04-22 sub_region ... 66 578.518188
67 2020-04-23 sub_region ... 67 578.516479
68 2020-04-24 sub_region ... 68 578.510986
69 2020-04-25 sub_region ... 69 578.503906
70 2020-04-26 sub_region ... 70 578.503906
71 2020-04-27 sub_region ... 71 578.500427
72 2020-04-28 sub_region ... 72 578.506470
73 2020-04-29 sub_region ... 73 578.516052
74 2020-04-30 sub_region ... 74 578.512817
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/media/plotly/test_plot_20_ae1f0e4f.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ock9vc16
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173133-ock9vc16/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ock9vc16 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: h21b0f2d with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 1 use_mask: True wandb: Agent Started Run: h21b0f2d
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/h21b0f2d INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmgyMWIwZjJkOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
15.545928395004012
The number of items in train is:
28
The loss for epoch 0
0.5552117283930004
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/media/graph/graph_0_summary_d7c91db2.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/media/graph
The running loss is: 21.661523299873807 The number of items in train is: 28 The loss for epoch 1 0.7736258321383502 The running loss is: 14.07579830638133 The number of items in train is: 28 The loss for epoch 2 0.5027070823707618
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-history.jsonl
The running loss is: 11.572435699170455 The number of items in train is: 28 The loss for epoch 3 0.4133012749703734 1 The running loss is: 11.783770642941818 The number of items in train is: 28 The loss for epoch 4 0.42084895153363633 The running loss is: 9.61011728970334 The number of items in train is: 28 The loss for epoch 5 0.3432184746322621 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-history.jsonl
The running loss is: 13.171849727863446 The number of items in train is: 28 The loss for epoch 6 0.4704232045665516 The running loss is: 9.562567624612711 The number of items in train is: 28 The loss for epoch 7 0.34152027230759685 1 The running loss is: 11.004484111152124 The number of items in train is: 28 The loss for epoch 8 0.3930172896840044
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-history.jsonl
2 The running loss is: 10.858391232584836 The number of items in train is: 28 The loss for epoch 9 0.38779968687802985 Data saved to: 28_May_202005_31PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['h21b0f2d']
Data saved to: 28_May_202005_31PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 856.106934
66 2020-04-22 sub_region ... 66 898.349731
67 2020-04-23 sub_region ... 67 857.475891
68 2020-04-24 sub_region ... 68 861.468872
69 2020-04-25 sub_region ... 69 885.428162
70 2020-04-26 sub_region ... 70 897.812500
71 2020-04-27 sub_region ... 71 894.293274
72 2020-04-28 sub_region ... 72 929.067383
73 2020-04-29 sub_region ... 73 878.263550
74 2020-04-30 sub_region ... 74 869.549072
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/media/plotly/test_plot_20_314de4b9.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: h21b0f2d
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173154-h21b0f2d/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: h21b0f2d INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: j5xzumca with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 1 use_mask: False wandb: Agent Started Run: j5xzumca
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/j5xzumca INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmo1eHp1bWNhOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
15.36638443195261
The number of items in train is:
28
The loss for epoch 0
0.5487994439983075
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/media/graph/graph_0_summary_751e4c9c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/media
The running loss is: 19.380546062253416 The number of items in train is: 28 The loss for epoch 1 0.6921623593661934 The running loss is: 13.36081693135202 The number of items in train is: 28 The loss for epoch 2 0.4771720332625721
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-history.jsonl
The running loss is: 11.953934513032436 The number of items in train is: 28 The loss for epoch 3 0.4269262326083013 1 The running loss is: 12.163511937658768 The number of items in train is: 28 The loss for epoch 4 0.43441114063067027 The running loss is: 9.378338906506542 The number of items in train is: 28 The loss for epoch 5 0.33494067523237653 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-history.jsonl
The running loss is: 12.704894061491359 The number of items in train is: 28 The loss for epoch 6 0.4537462164818343 The running loss is: 11.437989893951453 The number of items in train is: 28 The loss for epoch 7 0.4084996390696948 1 The running loss is: 11.367679933318868 The number of items in train is: 28 The loss for epoch 8 0.40598856904710245
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-history.jsonl
The running loss is: 11.42585400538519 The number of items in train is: 28 The loss for epoch 9 0.40806621447804253 1 Data saved to: 28_May_202005_32PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_32PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['j5xzumca'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 802.753906
66 2020-04-22 sub_region ... 66 877.084839
67 2020-04-23 sub_region ... 67 830.681274
68 2020-04-24 sub_region ... 68 816.030273
69 2020-04-25 sub_region ... 69 821.921509
70 2020-04-26 sub_region ... 70 826.182373
71 2020-04-27 sub_region ... 71 846.830322
72 2020-04-28 sub_region ... 72 856.710266
73 2020-04-29 sub_region ... 73 857.412720
74 2020-04-30 sub_region ... 74 841.012085
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/media/plotly/test_plot_20_e30096eb.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: j5xzumca
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173210-j5xzumca/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: j5xzumca INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: hhsbtt8o with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 2 use_mask: True wandb: Agent Started Run: hhsbtt8o
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/hhsbtt8o INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmhoc2J0dDhvOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
10.473457891494036
The number of items in train is:
27
The loss for epoch 0
0.38790584783311244
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/media/graph/graph_0_summary_bd7e2cbc.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/media/graph
The running loss is: 12.50141921825707 The number of items in train is: 27 The loss for epoch 1 0.4630155266021137 1 The running loss is: 12.783501834142953 The number of items in train is: 27 The loss for epoch 2 0.47346303089418346
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-history.jsonl
The running loss is: 5.422599470242858 The number of items in train is: 27 The loss for epoch 3 0.20083701741640214 1 The running loss is: 5.399664411786944 The number of items in train is: 27 The loss for epoch 4 0.19998757080692384 The running loss is: 4.413807349279523 The number of items in train is: 27 The loss for epoch 5 0.16347434626961196
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-history.jsonl
The running loss is: 4.564219214487821 The number of items in train is: 27 The loss for epoch 6 0.16904515609214152 The running loss is: 4.552914118394256 The number of items in train is: 27 The loss for epoch 7 0.16862644882941688 1 The running loss is: 4.2418790506199 The number of items in train is: 27 The loss for epoch 8 0.15710663150444074 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-history.jsonl
The running loss is: 4.331287162844092 The number of items in train is: 27 The loss for epoch 9 0.16041804306829968 Data saved to: 28_May_202005_32PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_32PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['hhsbtt8o'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 839.469238
66 2020-04-22 sub_region ... 66 751.845520
67 2020-04-23 sub_region ... 67 796.254761
68 2020-04-24 sub_region ... 68 840.047058
69 2020-04-25 sub_region ... 69 856.757812
70 2020-04-26 sub_region ... 70 857.516724
71 2020-04-27 sub_region ... 71 805.222778
72 2020-04-28 sub_region ... 72 846.005188
73 2020-04-29 sub_region ... 73 746.322815
74 2020-04-30 sub_region ... 74 799.226501
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/media/plotly/test_plot_20_908a465b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: hhsbtt8o
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173221-hhsbtt8o/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: hhsbtt8o INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 1pjcq596 with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 2 use_mask: False wandb: Agent Started Run: 1pjcq596
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/1pjcq596 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjFwamNxNTk2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
10.56319503299892
The number of items in train is:
27
The loss for epoch 0
0.39122944566662665
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/media/graph/graph_0_summary_20b617bc.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/media/graph
The running loss is: 12.365171766839921 The number of items in train is: 27 The loss for epoch 1 0.45796932469777485 1 The running loss is: 10.063402196858078 The number of items in train is: 27 The loss for epoch 2 0.37271859988363254
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-summary.json
The running loss is: 5.191733076237142 The number of items in train is: 27 The loss for epoch 3 0.19228641023100526 1 The running loss is: 5.048446481116116 The number of items in train is: 27 The loss for epoch 4 0.1869794993005969 The running loss is: 4.6334264650940895 The number of items in train is: 27 The loss for epoch 5 0.17160838759607738 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-summary.json
The running loss is: 4.809368921909481 The number of items in train is: 27 The loss for epoch 6 0.17812477488553635 The running loss is: 4.469904407858849 The number of items in train is: 27 The loss for epoch 7 0.16555201510588327 1 The running loss is: 4.4312631585635245 The number of items in train is: 27 The loss for epoch 8 0.164120857724575
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-summary.json
The running loss is: 4.365964470198378 The number of items in train is: 27 The loss for epoch 9 0.1617023877851251 1 Data saved to: 28_May_202005_32PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_32PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['1pjcq596'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 820.661011
66 2020-04-22 sub_region ... 66 731.493591
67 2020-04-23 sub_region ... 67 777.978394
68 2020-04-24 sub_region ... 68 817.230530
69 2020-04-25 sub_region ... 69 833.734131
70 2020-04-26 sub_region ... 70 830.960205
71 2020-04-27 sub_region ... 71 781.981934
72 2020-04-28 sub_region ... 72 821.894836
73 2020-04-29 sub_region ... 73 732.720215
74 2020-04-30 sub_region ... 74 782.010132
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/media/plotly/test_plot_20_a6760719.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 1pjcq596
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173237-1pjcq596/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 1pjcq596 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: pwxzkbkd with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 3 use_mask: True wandb: Agent Started Run: pwxzkbkd
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/pwxzkbkd INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnB3eHprYmtkOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
22.41447990387678
The number of items in train is:
27
The loss for epoch 0
0.8301659223658068
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/media/graph/graph_0_summary_4da48f4f.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/media
The running loss is: 20.35543255507946 The number of items in train is: 27 The loss for epoch 1 0.7539049094473874 1 The running loss is: 19.27651271224022 The number of items in train is: 27 The loss for epoch 2 0.7139449152681563
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-summary.json
The running loss is: 12.351722501218319 The number of items in train is: 27 The loss for epoch 3 0.4574712037488266 1 The running loss is: 9.813126027584076 The number of items in train is: 27 The loss for epoch 4 0.36344911213274356 The running loss is: 6.200845863670111 The number of items in train is: 27 The loss for epoch 5 0.2296609579137078 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-summary.json
The running loss is: 5.99026171118021 The number of items in train is: 27 The loss for epoch 6 0.2218615448585263 The running loss is: 5.51112406142056 The number of items in train is: 27 The loss for epoch 7 0.20411570597853926 1 The running loss is: 5.962472440674901 The number of items in train is: 27 The loss for epoch 8 0.22083231261758893
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-summary.json
The running loss is: 5.207641955465078 The number of items in train is: 27 The loss for epoch 9 0.1928756279801881 1 Data saved to: 28_May_202005_32PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['pwxzkbkd']
Data saved to: 28_May_202005_32PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 731.740784
66 2020-04-22 sub_region ... 66 738.606201
67 2020-04-23 sub_region ... 67 737.777405
68 2020-04-24 sub_region ... 68 726.210022
69 2020-04-25 sub_region ... 69 719.908691
70 2020-04-26 sub_region ... 70 719.150635
71 2020-04-27 sub_region ... 71 720.189941
72 2020-04-28 sub_region ... 72 731.834473
73 2020-04-29 sub_region ... 73 735.288940
74 2020-04-30 sub_region ... 74 731.156189
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/media/plotly/test_plot_20_3857be08.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: pwxzkbkd
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173253-pwxzkbkd/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: pwxzkbkd INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 0laz5778 with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 3 use_mask: False wandb: Agent Started Run: 0laz5778
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/0laz5778 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjBsYXo1Nzc4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
22.289740599691868
The number of items in train is:
27
The loss for epoch 0
0.8255459481367359
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/media/graph/graph_0_summary_42d3d77c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/media
The running loss is: 20.721937209367752 The number of items in train is: 27 The loss for epoch 1 0.7674791559025094 1 The running loss is: 19.501514554023743 The number of items in train is: 27 The loss for epoch 2 0.7222783168156942
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-summary.json
The running loss is: 13.022847175598145 The number of items in train is: 27 The loss for epoch 3 0.4823276731703017 1 The running loss is: 9.215275160968304 The number of items in train is: 27 The loss for epoch 4 0.3413064874432705 2 The running loss is: 6.453644126653671 The number of items in train is: 27 The loss for epoch 5 0.23902385654272856
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-summary.json
The running loss is: 6.3195782247930765 The number of items in train is: 27 The loss for epoch 6 0.23405845277011395 1 The running loss is: 5.538634759373963 The number of items in train is: 27 The loss for epoch 7 0.20513462071755417 The running loss is: 6.0211376789957285 The number of items in train is: 27 The loss for epoch 8 0.22300509922206402 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-summary.json
The running loss is: 5.4065384501591325 The number of items in train is: 27 The loss for epoch 9 0.2002421648207086 Data saved to: 28_May_202005_33PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_33PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['0laz5778']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-history.jsonl
CSV Path below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-summary.json
United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/config.yaml
torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 723.085754
66 2020-04-22 sub_region ... 66 731.862305
67 2020-04-23 sub_region ... 67 732.741089
68 2020-04-24 sub_region ... 68 723.526733
69 2020-04-25 sub_region ... 69 715.763306
70 2020-04-26 sub_region ... 70 713.154541
71 2020-04-27 sub_region ... 71 712.077576
72 2020-04-28 sub_region ... 72 725.158813
73 2020-04-29 sub_region ... 73 729.568176
74 2020-04-30 sub_region ... 74 726.045166
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/media/plotly/test_plot_20_967a32da.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 0laz5778
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173304-0laz5778/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 0laz5778 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ik8pyfox with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 4 use_mask: True wandb: Agent Started Run: ik8pyfox
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ik8pyfox INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmlrOHB5Zm94OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
17.69617347419262
The number of items in train is:
26
The loss for epoch 0
0.6806220566997161
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/media/graph/graph_0_summary_71492d1d.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/media
The running loss is: 12.942448504269123 The number of items in train is: 26 The loss for epoch 1 0.4977864809334278 1 The running loss is: 11.007499657571316 The number of items in train is: 26 The loss for epoch 2 0.4233653714450506 2 The running loss is: 7.731042757630348 The number of items in train is: 26 The loss for epoch 3 0.297347798370398
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-summary.json
The running loss is: 4.677496463060379 The number of items in train is: 26 The loss for epoch 4 0.1799037101177069 1 The running loss is: 3.537908681668341 The number of items in train is: 26 The loss for epoch 5 0.13607341083339775 The running loss is: 3.6070269970223308 The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-history.jsonl
26 The loss for epoch 6
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-summary.json
0.13873180757778195 1 The running loss is: 5.469477290287614 The number of items in train is: 26 The loss for epoch 7 0.21036451116490823 2 The running loss is: 4.373710255138576 The number of items in train is: 26 The loss for epoch 8 0.16821962519763753
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-summary.json
The running loss is: 4.385507521219552 The number of items in train is: 26 The loss for epoch 9 0.168673366200752 Data saved to: 28_May_202005_33PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_33PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['ik8pyfox']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-summary.json /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 641.062012
66 2020-04-22 sub_region ... 66 649.469788
67 2020-04-23 sub_region ... 67 662.655884
68 2020-04-24 sub_region ... 68 663.000610
69 2020-04-25 sub_region ... 69 652.813110
70 2020-04-26 sub_region ... 70 647.735291
71 2020-04-27 sub_region ... 71 634.131409
72 2020-04-28 sub_region ... 72 641.305664
73 2020-04-29 sub_region ... 73 641.875549
74 2020-04-30 sub_region ... 74 656.309570
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/media/plotly/test_plot_20_7320596b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ik8pyfox
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173315-ik8pyfox/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ik8pyfox INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 86vfy9ok with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 4 use_mask: False wandb: Agent Started Run: 86vfy9ok
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/86vfy9ok INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjg2dmZ5OW9rOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
17.936274334788322
The number of items in train is:
26
The loss for epoch 0
0.6898567051841662
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/media/graph/graph_0_summary_52568ab3.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/media/graph
The running loss is: 12.563346169888973 The number of items in train is: 26 The loss for epoch 1 0.48320562191880667 1 The running loss is: 10.63600341975689 The number of items in train is: 26 The loss for epoch 2 0.4090770546060342 2 The running loss is: 7.032686905935407 The number of items in train is: 26 The loss for epoch 3 0.27048795792059255
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-summary.json
The running loss is: 4.1291360929608345 The number of items in train is: 26 The loss for epoch 4 0.1588129266523398 1 The running loss is: 3.5215289955958724 The number of items in train is: 26 The loss for epoch 5 0.13544342290753356 The running loss is: 4.55407845415175 The number of items in train is: 26 The loss for epoch 6 0.17515686362122113
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-summary.json
1 The running loss is: 3.83646054379642 The number of items in train is: 26 The loss for epoch 7 0.14755617476140076 2 The running loss is: 4.684140918310732 The number of items in train is: 26 The loss for epoch 8 0.18015926608887428
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-summary.json
The running loss is: 7.284909620881081 The number of items in train is: 26 The loss for epoch 9 0.28018883157234925 1 Data saved to: 28_May_202005_33PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_33PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['86vfy9ok'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 811.324219
66 2020-04-22 sub_region ... 66 801.534241
67 2020-04-23 sub_region ... 67 837.117920
68 2020-04-24 sub_region ... 68 854.320984
69 2020-04-25 sub_region ... 69 834.952698
70 2020-04-26 sub_region ... 70 831.260376
71 2020-04-27 sub_region ... 71 782.874146
72 2020-04-28 sub_region ... 72 832.120972
73 2020-04-29 sub_region ... 73 785.114258
74 2020-04-30 sub_region ... 74 817.378418
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/media/plotly/test_plot_20_d2368e1b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 86vfy9ok
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173326-86vfy9ok/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 86vfy9ok INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: xlla4e7k with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 5 use_mask: True wandb: Agent Started Run: xlla4e7k
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/xlla4e7k INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnhsbGE0ZTdrOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
16.614094384014606
The number of items in train is:
26
The loss for epoch 0
0.639003630154408
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/media/graph/graph_0_summary_d7ce5ecf.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/media/graph
The running loss is: 12.784377932548523 The number of items in train is: 26 The loss for epoch 1 0.49170684355955857 1 The running loss is: 10.317914854735136 The number of items in train is: 26 The loss for epoch 2 0.3968428790282745
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-summary.json
The running loss is: 8.098226331174374 The number of items in train is: 26 The loss for epoch 3 0.31147024350670666 1 The running loss is: 5.721522863255814 The number of items in train is: 26 The loss for epoch 4 0.22005857166368514 The running loss is: 4.689897304400802 The number of items in train is: 26 The loss for epoch 5 0.180380665553877
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-summary.json
The running loss is: 6.495792439207435 The number of items in train is: 26 The loss for epoch 6 0.24983817073874748 1 The running loss is: 5.418599771335721 The number of items in train is: 26 The loss for epoch 7 0.20840768351291233 The running loss is: 4.195066409185529 The number of items in train is: 26 The loss for epoch 8 0.16134870804559726 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-summary.json
The running loss is: 5.375888114795089 The number of items in train is: 26 The loss for epoch 9 0.2067649274921188 2 Data saved to: 28_May_202005_33PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_33PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['xlla4e7k'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 730.216797
66 2020-04-22 sub_region ... 66 741.966797
67 2020-04-23 sub_region ... 67 738.005005
68 2020-04-24 sub_region ... 68 724.475220
69 2020-04-25 sub_region ... 69 719.175964
70 2020-04-26 sub_region ... 70 717.805908
71 2020-04-27 sub_region ... 71 714.661133
72 2020-04-28 sub_region ... 72 731.062378
73 2020-04-29 sub_region ... 73 725.214966
74 2020-04-30 sub_region ... 74 723.802612
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/media/plotly/test_plot_20_54bf2aea.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: xlla4e7k
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173337-xlla4e7k/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: xlla4e7k INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: nlztwjca with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 1 out_seq_length: 5 use_mask: False wandb: Agent Started Run: nlztwjca
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/nlztwjca INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm5senR3amNhOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
16.377284348011017
The number of items in train is:
26
The loss for epoch 0
0.6298955518465775
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/media/graph/graph_0_summary_92ff03f7.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/media
The running loss is: 14.166790530085564 The number of items in train is: 26 The loss for epoch 1 0.5448765588494447 The running loss is: 9.428598899394274 The number of items in train is: 26 The loss for epoch 2 0.3626384192074721 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/wandb-summary.json
The running loss is: 6.6330753564834595 The number of items in train is: 26 The loss for epoch 3 0.2551182829416715 2 The running loss is: 5.67338194232434 The number of items in train is: 26 The loss for epoch 4 0.2182069977817054 3 Stopping model now Data saved to: 28_May_202005_33PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_33PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 339.937469
66 2020-04-22 sub_region ... 66 348.662567
67 2020-04-23 sub_region ... 67 321.414612
68 2020-04-24 sub_region ... 68 302.783997
69 2020-04-25 sub_region ... 69 324.099854
70 2020-04-26 sub_region ... 70 317.279297
71 2020-04-27 sub_region ... 71 371.108948
72 2020-04-28 sub_region ... 72 297.625122
73 2020-04-29 sub_region ... 73 384.647125
74 2020-04-30 sub_region ... 74 332.773010
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
INFO:wandb.wandb_agent:Running runs: ['nlztwjca']
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/media/plotly/test_plot_10_807ad6e3.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: nlztwjca
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/media/plotly/test_plot_all_11_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173353-nlztwjca/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: nlztwjca INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 2ra81s8w with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 1 use_mask: True wandb: Agent Started Run: 2ra81s8w
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/2ra81s8w INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjJyYTgxczh3OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/media/graph/graph_0_summary_737450e3.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/media
The running loss is: 16.114518700167537 The number of items in train is: 28 The loss for epoch 0 0.5755185250059834 The running loss is: 40.93923968449235 The number of items in train is: 28 The loss for epoch 1 1.462115703017584
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-history.jsonl
The running loss is: 17.72650630073622 The number of items in train is: 28 The loss for epoch 2 0.6330895107405793 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-history.jsonl
The running loss is: 13.82603452168405 The number of items in train is: 28 The loss for epoch 3 0.4937869472030018 The running loss is: 12.235369089990854 The number of items in train is: 28 The loss for epoch 4 0.43697746749967337
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-history.jsonl
The running loss is: 12.42129836557433 The number of items in train is: 28 The loss for epoch 5 0.4436177987705118 1
INFO:wandb.wandb_agent:Running runs: ['2ra81s8w']
The running loss is: 11.593332045944408 The number of items in train is: 28 The loss for epoch 6 0.4140475730694431
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-history.jsonl
The running loss is: 12.022986069088802 The number of items in train is: 28 The loss for epoch 7 0.42939235961031436 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-history.jsonl
The running loss is: 11.050711158488411 The number of items in train is: 28 The loss for epoch 8 0.3946682556603004 2 The running loss is: 10.95676115965398 The number of items in train is: 28 The loss for epoch 9 0.3913128985590707 3 Stopping model now Data saved to: 28_May_202005_34PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-history.jsonl
Data saved to: 28_May_202005_34PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 837.403687
66 2020-04-22 sub_region ... 66 830.898865
67 2020-04-23 sub_region ... 67 824.932678
68 2020-04-24 sub_region ... 68 836.498291
69 2020-04-25 sub_region ... 69 841.817627
70 2020-04-26 sub_region ... 70 846.289368
71 2020-04-27 sub_region ... 71 830.351929
72 2020-04-28 sub_region ... 72 852.223999
73 2020-04-29 sub_region ... 73 823.185059
74 2020-04-30 sub_region ... 74 832.774414
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 2ra81s8w
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/media/plotly/test_plot_20_754d0a08.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173404-2ra81s8w/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 2ra81s8w INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: oouro82l with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 1 use_mask: False wandb: Agent Started Run: oouro82l
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/oouro82l INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm9vdXJvODJsOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/media/graph/graph_0_summary_607f15ec.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/media/graph
The running loss is: 15.39689282909967 The number of items in train is: 28 The loss for epoch 0 0.5498890296107025 The running loss is: 44.767726235091686 The number of items in train is: 28 The loss for epoch 1 1.5988473655389888
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/wandb-history.jsonl
The running loss is: 22.245736518874764 The number of items in train is: 28 The loss for epoch 2 0.794490589959813 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/wandb-history.jsonl
The running loss is: 15.605747589841485 The number of items in train is: 28 The loss for epoch 3 0.5573481282086244 2 The running loss is: 18.130321642383933 The number of items in train is: 28 The loss for epoch 4 0.6475114872279976 3 Stopping model now Data saved to: 28_May_202005_34PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/wandb-history.jsonl
Data saved to: 28_May_202005_34PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['oouro82l']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 410.513519
66 2020-04-22 sub_region ... 66 402.391907
67 2020-04-23 sub_region ... 67 409.347412
68 2020-04-24 sub_region ... 68 416.087708
69 2020-04-25 sub_region ... 69 416.266327
70 2020-04-26 sub_region ... 70 415.997223
71 2020-04-27 sub_region ... 71 405.508575
72 2020-04-28 sub_region ... 72 415.981689
73 2020-04-29 sub_region ... 73 393.018555
74 2020-04-30 sub_region ... 74 405.671448
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: oouro82l
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/media/plotly/test_plot_10_d4fb56b9.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/media/plotly/test_plot_all_11_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173420-oouro82l/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: oouro82l INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: t6rqbam0 with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 2 use_mask: True wandb: Agent Started Run: t6rqbam0
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/t6rqbam0 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnQ2cnFiYW0wOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/media/graph/graph_0_summary_55c20091.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/media
The running loss is: 10.718430198729038 The number of items in train is: 27 The loss for epoch 0 0.3969788962492236 The running loss is: 28.698574017733335 The number of items in train is: 27 The loss for epoch 1 1.0629101488049384
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-summary.json
The running loss is: 16.018407717347145 The number of items in train is: 27 The loss for epoch 2 0.5932743599017462 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-summary.json
The running loss is: 10.534781254827976 The number of items in train is: 27 The loss for epoch 3 0.39017708351214725 2 The running loss is: 5.385364955291152 The number of items in train is: 27 The loss for epoch 4 0.1994579613070797
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-summary.json
The running loss is: 7.625352939590812 The number of items in train is: 27 The loss for epoch 5 0.28242047924410413 1
INFO:wandb.wandb_agent:Running runs: ['t6rqbam0']
The running loss is: 6.7238668041536584 The number of items in train is: 27 The loss for epoch 6 0.2490321038575429
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-summary.json
The running loss is: 7.8772421116009355 The number of items in train is: 27 The loss for epoch 7 0.29174970783707166
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-summary.json
The running loss is: 7.317102946341038 The number of items in train is: 27 The loss for epoch 8 0.27100381282744584 1 The running loss is: 5.420254968106747 The number of items in train is: 27 The loss for epoch 9 0.20075018400395359 Data saved to: 28_May_202005_34PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-summary.json
Data saved to: 28_May_202005_34PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 779.043823
66 2020-04-22 sub_region ... 66 778.755005
67 2020-04-23 sub_region ... 67 778.881592
68 2020-04-24 sub_region ... 68 779.025024
69 2020-04-25 sub_region ... 69 779.112366
70 2020-04-26 sub_region ... 70 779.083008
71 2020-04-27 sub_region ... 71 778.987671
72 2020-04-28 sub_region ... 72 778.948608
73 2020-04-29 sub_region ... 73 778.723877
74 2020-04-30 sub_region ... 74 778.891602
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/media/plotly/test_plot_20_03adc1c1.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/media/plotly
wandb: Agent Finished Run: t6rqbam0
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173434-t6rqbam0/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: t6rqbam0 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: fzr1hshg with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 2 use_mask: False wandb: Agent Started Run: fzr1hshg
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/fzr1hshg INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmZ6cjFoc2hnOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/media/graph/graph_0_summary_dcf0e700.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/media/graph
The running loss is: 10.668751269578934 The number of items in train is: 27 The loss for epoch 0 0.3951389359103309 The running loss is: 28.521158926188946 The number of items in train is: 27 The loss for epoch 1 1.0563392194884795
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-summary.json
The running loss is: 14.176970317959785 The number of items in train is: 27 The loss for epoch 2 0.5250729747392513 The running loss is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-summary.json
6.591888493858278 The number of items in train is: 27 The loss for epoch 3 0.24414401829104732 1 The running loss is: 7.487370473332703 The number of items in train is: 27 The loss for epoch 4 0.27731001753084084
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-summary.json
The running loss is: 5.26125293970108 The number of items in train is: 27 The loss for epoch 5 0.1948612199889289
INFO:wandb.wandb_agent:Running runs: ['fzr1hshg']
The running loss is: 6.058157247491181 The number of items in train is: 27 The loss for epoch 6 0.22437619435152523 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-summary.json
The running loss is: 4.886977478861809 The number of items in train is: 27 The loss for epoch 7 0.18099916588377069
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-summary.json
The running loss is: 8.049873136915267 The number of items in train is: 27 The loss for epoch 8 0.29814344951538024 1 The running loss is: 6.706313902512193 The number of items in train is: 27 The loss for epoch 9 0.24838199638934047 2 Data saved to: 28_May_202005_34PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-summary.json
Data saved to: 28_May_202005_34PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 755.005249
66 2020-04-22 sub_region ... 66 751.443115
67 2020-04-23 sub_region ... 67 753.844849
68 2020-04-24 sub_region ... 68 755.624390
69 2020-04-25 sub_region ... 69 755.841370
70 2020-04-26 sub_region ... 70 755.883606
71 2020-04-27 sub_region ... 71 752.995850
72 2020-04-28 sub_region ... 72 755.255066
73 2020-04-29 sub_region ... 73 751.328247
74 2020-04-30 sub_region ... 74 753.881714
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: fzr1hshg
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/media/plotly/test_plot_20_9c0dfd36.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173451-fzr1hshg/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: fzr1hshg INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 1rhqvm3o with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 3 use_mask: True wandb: Agent Started Run: 1rhqvm3o
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/1rhqvm3o INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjFyaHF2bTNvOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/media/graph/graph_0_summary_a757ef87.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/media/graph
The running loss is: 23.849574618041515 The number of items in train is: 27 The loss for epoch 0 0.883317578445982 The running loss is: 20.568836465477943 The number of items in train is: 27 The loss for epoch 1 0.7618087579806646
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-history.jsonl
The running loss is: 19.27038937062025 The number of items in train is: 27 The loss for epoch 2 0.7137181248377871
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-history.jsonl
The running loss is: 17.25364562869072 The number of items in train is: 27 The loss for epoch 3 0.6390239121737303 1 The running loss is: 14.78194186091423 The number of items in train is: 27 The loss for epoch 4 0.5474793281820085
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-history.jsonl
The running loss is: 10.99204146116972 The number of items in train is: 27 The loss for epoch 5 0.40711264670998965 1
INFO:wandb.wandb_agent:Running runs: ['1rhqvm3o']
The running loss is: 34.4778664149344 The number of items in train is: 27 The loss for epoch 6 1.2769580153679405
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-history.jsonl
The running loss is: 18.340780273079872 The number of items in train is: 27 The loss for epoch 7 0.6792881582622174 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-history.jsonl
The running loss is: 18.76011623442173 The number of items in train is: 27 The loss for epoch 8 0.6948191197933974 2 The running loss is: 18.743604812771082 The number of items in train is: 27 The loss for epoch 9 0.6942075856581882 Data saved to: 28_May_202005_35PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_35PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 351.287323
66 2020-04-22 sub_region ... 66 351.352722
67 2020-04-23 sub_region ... 67 351.261353
68 2020-04-24 sub_region ... 68 351.227020
69 2020-04-25 sub_region ... 69 351.256317
70 2020-04-26 sub_region ... 70 351.258667
71 2020-04-27 sub_region ... 71 351.388000
72 2020-04-28 sub_region ... 72 351.195374
73 2020-04-29 sub_region ... 73 351.413818
74 2020-04-30 sub_region ... 74 351.286987
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/media/plotly/test_plot_20_3234c249.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 1rhqvm3o
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173508-1rhqvm3o/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 1rhqvm3o INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: hb05ymdu with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 3 use_mask: False wandb: Agent Started Run: hb05ymdu
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/hb05ymdu INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmhiMDV5bWR1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/media/graph/graph_0_summary_69776693.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/media
The running loss is: 23.785254307091236 The number of items in train is: 27 The loss for epoch 0 0.8809353447070828 The running loss is: 20.365187123417854 The number of items in train is: 27 The loss for epoch 1 0.7542661897562168
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-summary.json
The running loss is: 19.407820589840412 The number of items in train is: 27 The loss for epoch 2 0.7188081699940894
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-summary.json
The running loss is: 16.047366067767143 The number of items in train is: 27 The loss for epoch 3 0.594346891398783 1 The running loss is: 16.049381844699383 The number of items in train is: 27 The loss for epoch 4 0.5944215498036809
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-summary.json
The running loss is: 8.695411691442132 The number of items in train is: 27 The loss for epoch 5 0.3220522848682271 1
INFO:wandb.wandb_agent:Running runs: ['hb05ymdu']
The running loss is: 11.989098764955997 The number of items in train is: 27 The loss for epoch 6 0.4440406949983703
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-summary.json
The running loss is: 12.975145380944014 The number of items in train is: 27 The loss for epoch 7 0.48056094003496347 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-history.jsonl
The running loss is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-summary.json
6.981494817882776 The number of items in train is: 27 The loss for epoch 8 0.25857388214380655 The running loss is: 7.491051498800516 The number of items in train is: 27 The loss for epoch 9 0.27744635180742655 1 Data saved to: 28_May_202005_35PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-summary.json
Data saved to: 28_May_202005_35PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 709.696960
66 2020-04-22 sub_region ... 66 709.161499
67 2020-04-23 sub_region ... 67 708.643188
68 2020-04-24 sub_region ... 68 708.597656
69 2020-04-25 sub_region ... 69 708.908020
70 2020-04-26 sub_region ... 70 708.755493
71 2020-04-27 sub_region ... 71 709.836182
72 2020-04-28 sub_region ... 72 708.476990
73 2020-04-29 sub_region ... 73 709.191528
74 2020-04-30 sub_region ... 74 708.709473
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/media/plotly/test_plot_20_509b02c0.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: hb05ymdu
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173524-hb05ymdu/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: hb05ymdu INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: a81xvfoe with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 4 use_mask: True wandb: Agent Started Run: a81xvfoe
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/a81xvfoe INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmE4MXh2Zm9lOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/media/graph/graph_0_summary_c9c3fc1e.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/media/graph
The running loss is: 25.31017677485943 The number of items in train is: 26 The loss for epoch 0 0.9734683374945934 The running loss is: 17.691114902496338 The number of items in train is: 26 The loss for epoch 1 0.6804274962498591 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-summary.json
The running loss is: 21.284134477376938 The number of items in train is: 26 The loss for epoch 2 0.8186205568221899 The running loss is: 14.811767756938934 The number of items in train is: 26 The loss for epoch 3 0.5696833752668821
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-summary.json
The running loss is: 12.015214286744595 The number of items in train is: 26 The loss for epoch 4 0.46212362641325366 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-summary.json
The running loss is: 10.353260762989521 The number of items in train is: 26 The loss for epoch 5 0.3982023370380585 2 The running loss is: 7.667264841496944 The number of items in train is: 26 The loss for epoch 6 0.2948948015960363 3 Stopping model now Data saved to: 28_May_202005_35PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['a81xvfoe'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-summary.json
Data saved to: 28_May_202005_35PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 318.850616
66 2020-04-22 sub_region ... 66 316.706177
67 2020-04-23 sub_region ... 67 320.013977
68 2020-04-24 sub_region ... 68 321.565765
69 2020-04-25 sub_region ... 69 321.352051
70 2020-04-26 sub_region ... 70 320.382446
71 2020-04-27 sub_region ... 71 318.602081
72 2020-04-28 sub_region ... 72 318.317017
73 2020-04-29 sub_region ... 73 318.346680
74 2020-04-30 sub_region ... 74 320.484741
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: a81xvfoe
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/media/plotly/test_plot_all_15_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/media/plotly/test_plot_14_b1c54b64.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173540-a81xvfoe/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: a81xvfoe INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: gg0yv0kt with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 4 use_mask: False wandb: Agent Started Run: gg0yv0kt
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/gg0yv0kt INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmdnMHl2MGt0OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/media/graph/graph_0_summary_6680c38e.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/media
The running loss is: 25.090075597167015 The number of items in train is: 26 The loss for epoch 0 0.9650029075833467 The running loss is: 17.45224541425705 The number of items in train is: 26 The loss for epoch 1 0.6712402082406558 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-history.jsonl
The running loss is: 21.34785506129265 The number of items in train is: 26 The loss for epoch 2 0.8210713485112557 The running loss is: 14.617094554007053 The number of items in train is: 26 The loss for epoch 3 0.5621959443848866
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-history.jsonl
1 The running loss is: 12.857044279575348 The number of items in train is: 26 The loss for epoch 4 0.4945017030605903
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-history.jsonl
The running loss is: 10.908861950039864 The number of items in train is: 26 The loss for epoch 5 0.4195716134630717 1 The running loss is: 7.671748608350754 The number of items in train is: 26 The loss for epoch 6 0.2950672541673367
INFO:wandb.wandb_agent:Running runs: ['gg0yv0kt'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-history.jsonl
The running loss is: 6.437005430459976 The number of items in train is: 26 The loss for epoch 7 0.2475771319407683 The running loss is: 7.865247845649719 The number of items in train is: 26 The loss for epoch 8 0.3025095325249892 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-history.jsonl
The running loss is: 5.2042678743600845 The number of items in train is: 26 The loss for epoch 9 0.2001641490138494 Data saved to: 28_May_202005_35PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_35PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 570.512756
66 2020-04-22 sub_region ... 66 571.143372
67 2020-04-23 sub_region ... 67 570.903015
68 2020-04-24 sub_region ... 68 570.691162
69 2020-04-25 sub_region ... 69 570.595093
70 2020-04-26 sub_region ... 70 570.591003
71 2020-04-27 sub_region ... 71 570.901611
72 2020-04-28 sub_region ... 72 570.663696
73 2020-04-29 sub_region ... 73 571.259827
74 2020-04-30 sub_region ... 74 571.026001
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: gg0yv0kt
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/media/plotly/test_plot_20_96e49c04.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173551-gg0yv0kt/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: gg0yv0kt INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 695h83kt with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 5 use_mask: True wandb: Agent Started Run: 695h83kt
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/695h83kt INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjY5NWg4M2t0OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/media/graph/graph_0_summary_f71ffa02.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/media/graph
The running loss is: 24.998789221048355 The number of items in train is: 26 The loss for epoch 0 0.9614918931172445 The running loss is: 19.62556444108486 The number of items in train is: 26 The loss for epoch 1 0.754829401580187 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-summary.json
The running loss is: 19.021592140197754 The number of items in train is: 26 The loss for epoch 2 0.7315996976999136 The running loss is: 13.53062754869461 The number of items in train is: 26 The loss for epoch 3 0.5204087518728696
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-summary.json
The running loss is: 11.939093425869942 The number of items in train is: 26 The loss for epoch 4 0.45919590099499774
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-summary.json
The running loss is: 12.370361641049385 The number of items in train is: 26 The loss for epoch 5 0.475783140040361 1 The running loss is: 10.105389013886452 The number of items in train is: 26 The loss for epoch 6 0.388668808226402
INFO:wandb.wandb_agent:Running runs: ['695h83kt'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-summary.json
The running loss is: 5.374523870646954 The number of items in train is: 26 The loss for epoch 7 0.20671245656334436 1 The running loss is: 4.875691753812134 The number of items in train is: 26 The loss for epoch 8 0.18752660591585132
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-summary.json
The running loss is: 7.123469600221142 The number of items in train is: 26 The loss for epoch 9 0.27397960000850546 1 Data saved to: 28_May_202005_36PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_36PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 697.734924
66 2020-04-22 sub_region ... 66 699.246887
67 2020-04-23 sub_region ... 67 698.447754
68 2020-04-24 sub_region ... 68 697.592102
69 2020-04-25 sub_region ... 69 697.349731
70 2020-04-26 sub_region ... 70 697.215332
71 2020-04-27 sub_region ... 71 698.537048
72 2020-04-28 sub_region ... 72 697.177002
73 2020-04-29 sub_region ... 73 699.459961
74 2020-04-30 sub_region ... 74 698.878723
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/media/plotly/test_plot_20_70b23860.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 695h83kt
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173607-695h83kt/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 695h83kt INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: cd3yn44i with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 2 out_seq_length: 5 use_mask: False wandb: Agent Started Run: cd3yn44i
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/cd3yn44i INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmNkM3luNDRpOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/media/graph/graph_0_summary_8d8da9f6.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/media
The running loss is: 24.827137678861618 The number of items in train is: 26 The loss for epoch 0 0.9548899107254468 The running loss is: 19.719864428043365 The number of items in train is: 26 The loss for epoch 1 0.758456324155514 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-history.jsonl
The running loss is: 18.63474379479885 The number of items in train is: 26 The loss for epoch 2 0.7167209151845712 The running loss is: 13.806637480854988 The number of items in train is: 26 The loss for epoch 3 0.5310245184944227
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-history.jsonl
1 The running loss is: 11.331828974187374 The number of items in train is: 26 The loss for epoch 4 0.4358395759302836
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-history.jsonl
The running loss is: 10.775343157351017 The number of items in train is: 26 The loss for epoch 5 0.4144362752827314 1 The running loss is: 10.32153557986021 The number of items in train is: 26 The loss for epoch 6 0.3969821376869312 2
INFO:wandb.wandb_agent:Running runs: ['cd3yn44i'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-history.jsonl
The running loss is: 7.037654765415937 The number of items in train is: 26 The loss for epoch 7 0.2706790294390745 The running loss is: 14.718151992186904 The number of items in train is: 26 The loss for epoch 8 0.5660827689302655 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-history.jsonl
The running loss is: 5.737017912324518 The number of items in train is: 26 The loss for epoch 9 0.22065453508940452 Data saved to: 28_May_202005_36PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_36PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 632.474121
66 2020-04-22 sub_region ... 66 634.329102
67 2020-04-23 sub_region ... 67 632.773071
68 2020-04-24 sub_region ... 68 631.713074
69 2020-04-25 sub_region ... 69 631.196228
70 2020-04-26 sub_region ... 70 631.151917
71 2020-04-27 sub_region ... 71 632.160889
72 2020-04-28 sub_region ... 72 631.541016
73 2020-04-29 sub_region ... 73 633.977478
74 2020-04-30 sub_region ... 74 632.605713
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/media/plotly/test_plot_20_a4b0535b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: cd3yn44i
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173623-cd3yn44i/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: cd3yn44i INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ew8yx2l9 with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 1 use_mask: True wandb: Agent Started Run: ew8yx2l9
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ew8yx2l9 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmV3OHl4Mmw5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/media/graph/graph_0_summary_17923498.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/media/graph
The running loss is: 16.056726850569248 The number of items in train is: 28 The loss for epoch 0 0.5734545303774732
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-summary.json
The running loss is: 46.208357490599155 The number of items in train is: 28 The loss for epoch 1 1.6502984818071127
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-summary.json
The running loss is: 29.032308392226696 The number of items in train is: 28 The loss for epoch 2 1.0368681568652391
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-summary.json
The running loss is: 28.841471418738365 The number of items in train is: 28 The loss for epoch 3 1.0300525506692273
INFO:wandb.wandb_agent:Running runs: ['ew8yx2l9'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-summary.json
The running loss is: 28.338371301069856 The number of items in train is: 28 The loss for epoch 4 1.0120846893239235 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-summary.json
The running loss is: 28.69572694809176 The number of items in train is: 28 The loss for epoch 5 1.0248473910032772 2 The running loss is: 19.646056853234768 The number of items in train is: 28 The loss for epoch 6 0.7016448876155275 3 Stopping model now
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-history.jsonl
Data saved to:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-summary.json
28_May_202005_36PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_36PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 429.154022
66 2020-04-22 sub_region ... 66 428.729492
67 2020-04-23 sub_region ... 67 429.150635
68 2020-04-24 sub_region ... 68 429.542450
69 2020-04-25 sub_region ... 69 429.516174
70 2020-04-26 sub_region ... 70 429.553772
71 2020-04-27 sub_region ... 71 428.790649
72 2020-04-28 sub_region ... 72 429.774963
73 2020-04-29 sub_region ... 73 428.176636
74 2020-04-30 sub_region ... 74 428.942596
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/media/plotly/test_plot_14_bdfadf79.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ew8yx2l9
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/media/plotly/test_plot_all_15_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173642-ew8yx2l9/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: ew8yx2l9 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: kqq3a8gv with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 1 use_mask: False wandb: Agent Started Run: kqq3a8gv
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/kqq3a8gv INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmtxcTNhOGd2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/media/graph/graph_0_summary_55480c76.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/media/graph
The running loss is: 13.879746047779918 The number of items in train is: 28 The loss for epoch 0 0.4957052159921399
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-summary.json
The running loss is: 50.16468261182308 The number of items in train is: 28 The loss for epoch 1 1.7915958075651102
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-summary.json
The running loss is: 28.752860818058252 The number of items in train is: 28 The loss for epoch 2 1.0268878863592232
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-summary.json
The running loss is: 28.75844579562545 The number of items in train is: 28 The loss for epoch 3 1.027087349843766
INFO:wandb.wandb_agent:Running runs: ['kqq3a8gv'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-summary.json
The running loss is: 28.228818399831653 The number of items in train is: 28 The loss for epoch 4 1.0081720857082732 1 The running loss is: 28.275303000351414
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-history.jsonl
The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-summary.json
28 The loss for epoch 5 1.0098322500125505 2 The running loss is: 19.328091056086123 The number of items in train is: 28 The loss for epoch 6 0.6902889662887901 3 Stopping model now Data saved to: 28_May_202005_37PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-summary.json
Data saved to: 28_May_202005_37PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 429.017181
66 2020-04-22 sub_region ... 66 428.986389
67 2020-04-23 sub_region ... 67 429.089111
68 2020-04-24 sub_region ... 68 429.224731
69 2020-04-25 sub_region ... 69 429.164246
70 2020-04-26 sub_region ... 70 429.191895
71 2020-04-27 sub_region ... 71 428.905151
72 2020-04-28 sub_region ... 72 429.181915
73 2020-04-29 sub_region ... 73 428.645477
74 2020-04-30 sub_region ... 74 428.955353
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: kqq3a8gv
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/media/plotly/test_plot_all_15_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/media/plotly/test_plot_14_562144b7.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173659-kqq3a8gv/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: kqq3a8gv INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: x7pfw8a7 with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 2 use_mask: True wandb: Agent Started Run: x7pfw8a7
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/x7pfw8a7 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOng3cGZ3OGE3OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/media/graph/graph_0_summary_d2057918.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/media
The running loss is: 9.515796359628439 The number of items in train is: 27 The loss for epoch 0 0.3524369022084607
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-history.jsonl
The running loss is: 31.958917625248432 The number of items in train is: 27 The loss for epoch 1 1.1836636157499418
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-history.jsonl
The running loss is: 17.400996170938015 The number of items in train is: 27 The loss for epoch 2 0.6444813396643709 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-history.jsonl
The running loss is: 17.490230657160282 The number of items in train is: 27 The loss for epoch 3 0.647786320635566
INFO:wandb.wandb_agent:Running runs: ['x7pfw8a7'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-history.jsonl
The running loss is: 17.88597560673952 The number of items in train is: 27 The loss for epoch 4 0.6624435409903526 The running loss is: 17.275350011885166 The number of items in train is: 27 The loss for epoch 5 0.6398277782179691
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-history.jsonl
The running loss is: 17.80729231238365 The number of items in train is: 27 The loss for epoch 6 0.6595293449030982
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-history.jsonl
The running loss is: 17.339848961681128 The number of items in train is: 27 The loss for epoch 7 0.6422166282104121
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-history.jsonl
The running loss is: 16.749763071537018 The number of items in train is: 27 The loss for epoch 8 0.6203615952421118
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-history.jsonl
The running loss is: 17.160026656463742 The number of items in train is: 27 The loss for epoch 9 0.6355565428319905 Data saved to: 28_May_202005_37PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_37PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 452.964355
66 2020-04-22 sub_region ... 66 448.373505
67 2020-04-23 sub_region ... 67 451.731903
68 2020-04-24 sub_region ... 68 454.205078
69 2020-04-25 sub_region ... 69 454.852173
70 2020-04-26 sub_region ... 70 454.945068
71 2020-04-27 sub_region ... 71 451.557831
72 2020-04-28 sub_region ... 72 453.470245
73 2020-04-29 sub_region ... 73 448.514465
74 2020-04-30 sub_region ... 74 451.503876
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: x7pfw8a7
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/media/plotly/test_plot_20_da2b038c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173715-x7pfw8a7/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: x7pfw8a7 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 83bi2wlq with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 2 use_mask: False wandb: Agent Started Run: 83bi2wlq
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/83bi2wlq INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjgzYmkyd2xxOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/media/graph/graph_0_summary_c1b67686.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/media/graph
The running loss is: 9.540501032024622 The number of items in train is: 27 The loss for epoch 0 0.353351890074986
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-history.jsonl
The running loss is: 32.01947431452572 The number of items in train is: 27 The loss for epoch 1 1.1859064560935453
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-history.jsonl
The running loss is: 17.433912321925163 The number of items in train is: 27 The loss for epoch 2 0.6457004563675987 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-history.jsonl
The running loss is: 17.372653737664223 The number of items in train is: 27 The loss for epoch 3 0.6434316199134897
INFO:wandb.wandb_agent:Running runs: ['83bi2wlq'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-history.jsonl
The running loss is: 17.937954049557447 The number of items in train is: 27 The loss for epoch 4 0.6643686685021277 The running loss is: 17.54516338557005 The number of items in train is: 27 The loss for epoch 5 0.649820866132224
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-history.jsonl
The running loss is: 17.718922704458237 The number of items in train is: 27 The loss for epoch 6 0.6562563964614162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-history.jsonl
The running loss is: 17.30660080164671 The number of items in train is: 27 The loss for epoch 7 0.6409852148758041
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-history.jsonl
The running loss is: 16.718926943838596 The number of items in train is: 27 The loss for epoch 8 0.6192195164384665
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-history.jsonl
The running loss is: 16.827449632808566 The number of items in train is: 27 The loss for epoch 9 0.6232388752892062 Data saved to: 28_May_202005_37PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_37PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 457.992615
66 2020-04-22 sub_region ... 66 454.040710
67 2020-04-23 sub_region ... 67 456.812286
68 2020-04-24 sub_region ... 68 459.140167
69 2020-04-25 sub_region ... 69 459.434723
70 2020-04-26 sub_region ... 70 459.551544
71 2020-04-27 sub_region ... 71 456.045074
72 2020-04-28 sub_region ... 72 458.655548
73 2020-04-29 sub_region ... 73 453.452576
74 2020-04-30 sub_region ... 74 456.783661
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 83bi2wlq
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/media/plotly/test_plot_20_c566c54f.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173736-83bi2wlq/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 83bi2wlq INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: hkxc7mnc with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 3 use_mask: True wandb: Agent Started Run: hkxc7mnc
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/hkxc7mnc INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmhreGM3bW5jOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/media/graph/graph_0_summary_037e9957.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/media/graph
The running loss is: 31.62208504229784 The number of items in train is: 27 The loss for epoch 0 1.17118833489992
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-summary.json
The running loss is: 19.408665716648102 The number of items in train is: 27 The loss for epoch 1 0.7188394709869668 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-summary.json
The running loss is: 20.407186299562454 The number of items in train is: 27 The loss for epoch 2 0.7558217147986094
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-summary.json
The running loss is: 20.506580412387848 The number of items in train is: 27 The loss for epoch 3 0.759502978236587
INFO:wandb.wandb_agent:Running runs: ['hkxc7mnc'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-summary.json
The running loss is: 20.30820331722498 The number of items in train is: 27 The loss for epoch 4 0.7521556784157399 The running loss is: 19.309663198888302 The number of items in train is: 27 The loss for epoch 5 0.7151727110699371 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-summary.json
The running loss is: 18.341759957373142 The number of items in train is: 27 The loss for epoch 6 0.679324442865672
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-summary.json
The running loss is: 17.737579837441444 The number of items in train is: 27 The loss for epoch 7 0.6569474013867201 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-summary.json
The running loss is: 19.28192499279976 The number of items in train is: 27 The loss for epoch 8 0.7141453701036947
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-summary.json
The running loss is: 18.758581429719925 The number of items in train is: 27 The loss for epoch 9 0.694762275174812 1 Data saved to: 28_May_202005_38PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_38PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 337.341278
66 2020-04-22 sub_region ... 66 337.345581
67 2020-04-23 sub_region ... 67 337.355377
68 2020-04-24 sub_region ... 68 337.360901
69 2020-04-25 sub_region ... 69 337.360565
70 2020-04-26 sub_region ... 70 337.355988
71 2020-04-27 sub_region ... 71 337.354065
72 2020-04-28 sub_region ... 72 337.343140
73 2020-04-29 sub_region ... 73 337.349548
74 2020-04-30 sub_region ... 74 337.355652
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/media/plotly/test_plot_20_e91edfe1.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: hkxc7mnc
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173758-hkxc7mnc/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: hkxc7mnc INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 684sh0pv with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 3 use_mask: False wandb: Agent Started Run: 684sh0pv
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/684sh0pv INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjY4NHNoMHB2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/media/graph/graph_0_summary_9c008942.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/media/graph
The running loss is: 31.686953715980053 The number of items in train is: 27 The loss for epoch 0 1.1735908783696316
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-history.jsonl
The running loss is: 19.326208144426346 The number of items in train is: 27 The loss for epoch 1 0.7157854868306054 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-history.jsonl
The running loss is: 20.44809077680111 The number of items in train is: 27 The loss for epoch 2 0.7573366954370782
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-history.jsonl
The running loss is: 20.70396327972412 The number of items in train is: 27 The loss for epoch 3 0.7668134548045971
INFO:wandb.wandb_agent:Running runs: ['684sh0pv'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-history.jsonl
The running loss is: 20.114036343991756 The number of items in train is: 27 The loss for epoch 4 0.7449643090367317 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-history.jsonl
The running loss is: 19.89861784130335 The number of items in train is: 27 The loss for epoch 5 0.7369858459741981 The running loss is: 18.967432893812656 The number of items in train is: 27 The loss for epoch 6 0.7024975145856539
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-history.jsonl
1 The running loss is: 19.17745415121317 The number of items in train is: 27 The loss for epoch 7 0.7102760796745619
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-history.jsonl
The running loss is: 19.25604397803545 The number of items in train is: 27 The loss for epoch 8 0.713186814001313
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-history.jsonl
The running loss is: 18.80481818318367 The number of items in train is: 27 The loss for epoch 9 0.6964747475253211 Data saved to: 28_May_202005_38PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-history.jsonl
Data saved to: 28_May_202005_38PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 353.958618
66 2020-04-22 sub_region ... 66 353.994934
67 2020-04-23 sub_region ... 67 353.966064
68 2020-04-24 sub_region ... 68 353.948608
69 2020-04-25 sub_region ... 69 353.950348
70 2020-04-26 sub_region ... 70 353.951172
71 2020-04-27 sub_region ... 71 353.985535
72 2020-04-28 sub_region ... 72 353.945618
73 2020-04-29 sub_region ... 73 354.007202
74 2020-04-30 sub_region ... 74 353.972107
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/media/plotly/test_plot_20_384740d4.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 684sh0pv
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173819-684sh0pv/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 684sh0pv INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: uys7i4es with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 4 use_mask: True wandb: Agent Started Run: uys7i4es
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/uys7i4es INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnV5czdpNGVzOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/media/graph/graph_0_summary_571ea964.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/media
The running loss is: 28.677166178822517 The number of items in train is: 26 The loss for epoch 0 1.1029679299547122
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-summary.json
The running loss is: 19.185990750789642 The number of items in train is: 26 The loss for epoch 1 0.737922721184217
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-summary.json
The running loss is: 20.213715583086014 The number of items in train is: 26 The loss for epoch 2 0.777450599349462
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-summary.json
The running loss is: 15.579263865947723 The number of items in train is: 26 The loss for epoch 3 0.5992024563826047 1
INFO:wandb.wandb_agent:Running runs: ['uys7i4es'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-summary.json
The running loss is: 15.614615187048912 The number of items in train is: 26 The loss for epoch 4 0.6005621225788043 The running loss is: 14.192182525992393 The number of items in train is: 26 The loss for epoch 5 0.5458531740766305
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-summary.json
1 The running loss is: 17.893233135342598 The number of items in train is: 26 The loss for epoch 6 0.6882012744362538
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-summary.json
The running loss is: 15.887526050209999 The number of items in train is: 26 The loss for epoch 7 0.6110586942388461
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-summary.json
The running loss is: 13.056910991668701 The number of items in train is: 26 The loss for epoch 8 0.50218888429495 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-summary.json
The running loss is: 10.211280450224876 The number of items in train is: 26 The loss for epoch 9 0.39274155577787984 2 Data saved to: 28_May_202005_38PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_38PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 409.355835
66 2020-04-22 sub_region ... 66 409.509674
67 2020-04-23 sub_region ... 67 409.471558
68 2020-04-24 sub_region ... 68 409.399261
69 2020-04-25 sub_region ... 69 409.354004
70 2020-04-26 sub_region ... 70 409.365173
71 2020-04-27 sub_region ... 71 409.424011
72 2020-04-28 sub_region ... 72 409.376404
73 2020-04-29 sub_region ... 73 409.552612
74 2020-04-30 sub_region ... 74 409.473022
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: uys7i4es
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/media/plotly/test_plot_20_eefb9a0f.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173842-uys7i4es/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: uys7i4es INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: p90u6d7b with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 4 use_mask: False wandb: Agent Started Run: p90u6d7b
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/p90u6d7b INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnA5MHU2ZDdiOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/media/graph/graph_0_summary_6a1615dc.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/media/graph
The running loss is: 28.61015349626541 The number of items in train is: 26 The loss for epoch 0 1.1003905190871313
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-history.jsonl
The running loss is: 19.129196166992188 The number of items in train is: 26 The loss for epoch 1 0.7357383141150842
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-history.jsonl
The running loss is: 20.223093792796135 The number of items in train is: 26 The loss for epoch 2 0.7778112997229283
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-history.jsonl
The running loss is: 15.592137679457664 The number of items in train is: 26 The loss for epoch 3 0.599697603056064 1
INFO:wandb.wandb_agent:Running runs: ['p90u6d7b']
The running loss is: 14.741975575685501 The number of items in train is: 26 The loss for epoch 4 0.5669990606032885
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-history.jsonl
The running loss is: 14.141681730747223 The number of items in train is: 26 The loss for epoch 5 0.5439108357979701 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-history.jsonl
The running loss is: 17.65192748606205 The number of items in train is: 26 The loss for epoch 6 0.6789202879254634
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-history.jsonl
The running loss is: 13.934842199087143 The number of items in train is: 26 The loss for epoch 7 0.5359554691956594 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-history.jsonl
The running loss is: 10.86246071755886 The number of items in train is: 26 The loss for epoch 8 0.4177869506753408
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-history.jsonl
The running loss is: 9.102634258568287 The number of items in train is: 26 The loss for epoch 9 0.3501013176372418 1 Data saved to: 28_May_202005_39PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_39PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-summary.json /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 475.105896
66 2020-04-22 sub_region ... 66 475.061249
67 2020-04-23 sub_region ... 67 475.085632
68 2020-04-24 sub_region ... 68 475.112213
69 2020-04-25 sub_region ... 69 475.114807
70 2020-04-26 sub_region ... 70 475.139618
71 2020-04-27 sub_region ... 71 475.079346
72 2020-04-28 sub_region ... 72 475.173126
73 2020-04-29 sub_region ... 73 475.043427
74 2020-04-30 sub_region ... 74 475.085358
[20 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/media/plotly/test_plot_20_114ca43c.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: p90u6d7b
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173903-p90u6d7b/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: p90u6d7b INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 6ujfa0hd with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 5 use_mask: True wandb: Agent Started Run: 6ujfa0hd
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/6ujfa0hd INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjZ1amZhMGhkOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/media/graph/graph_0_summary_ab8f9d5a.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/media/graph
The running loss is: 26.287698671221733 The number of items in train is: 26 The loss for epoch 0 1.0110653335085282
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-summary.json
The running loss is: 22.019511476159096 The number of items in train is: 26 The loss for epoch 1 0.8469042875445806
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-summary.json
The running loss is: 19.7527959048748 The number of items in train is: 26 The loss for epoch 2 0.7597229194182616
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-summary.json
The running loss is: 15.987582370638847 The number of items in train is: 26 The loss for epoch 3 0.6149070142553403 1
INFO:wandb.wandb_agent:Running runs: ['6ujfa0hd']
The running loss is: 15.31729331612587 The number of items in train is: 26 The loss for epoch 4 0.5891266660048411
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-summary.json
The running loss is: 13.650041669607162 The number of items in train is: 26 The loss for epoch 5 0.5250016026771985 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-summary.json
The running loss is: 13.348268389701843 The number of items in train is: 26 The loss for epoch 6 0.5133949380654556 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-summary.json
The running loss is: 11.254892647266388 The number of items in train is: 26 The loss for epoch 7 0.4328804864333226 3 Stopping model now Data saved to: 28_May_202005_39PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_39PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 347.929901
66 2020-04-22 sub_region ... 66 348.376862
67 2020-04-23 sub_region ... 67 348.171692
68 2020-04-24 sub_region ... 68 347.880554
69 2020-04-25 sub_region ... 69 347.761993
70 2020-04-26 sub_region ... 70 347.774658
71 2020-04-27 sub_region ... 71 348.046265
72 2020-04-28 sub_region ... 72 347.813934
73 2020-04-29 sub_region ... 73 348.664154
74 2020-04-30 sub_region ... 74 348.247253
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-history.jsonl
wandb: Agent Finished Run: 6ujfa0hd
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/media/plotly/test_plot_all_17_53c27da0.plotly.json
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/media/plotly/test_plot_16_3f181331.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173928-6ujfa0hd/wandb-events.jsonl INFO:wandb.wandb_agent:Cleaning up finished run: 6ujfa0hd INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: yrbu62av with config: batch_size: 2 forecast_history: 10 lr: 0.01 number_encoder_layers: 3 out_seq_length: 5 use_mask: False wandb: Agent Started Run: yrbu62av
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/yrbu62av INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnlyYnU2MmF2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 10
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 10
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 10
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 10
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/media/graph/graph_0_summary_a70512b4.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/media
The running loss is: 26.332314282655716 The number of items in train is: 26 The loss for epoch 0 1.0127813185636814
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-history.jsonl
The running loss is: 21.822422578930855 The number of items in train is: 26 The loss for epoch 1 0.8393239453434944
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-history.jsonl
The running loss is: 19.969776809215546 The number of items in train is: 26 The loss for epoch 2 0.7680683388159826
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-history.jsonl
The running loss is: 15.921830900013447 The number of items in train is: 26 The loss for epoch 3 0.6123781115389787 1
INFO:wandb.wandb_agent:Running runs: ['yrbu62av']
The running loss is: 15.494909837841988 The number of items in train is: 26 The loss for epoch 4 0.5959580706862303
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-history.jsonl
The running loss is: 14.208502128720284 The number of items in train is: 26 The loss for epoch 5 0.5464808511046263 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-history.jsonl
The running loss is: 16.151983082294464 The number of items in train is: 26 The loss for epoch 6 0.621230118549787 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-history.jsonl
The running loss is: 12.41115165501833 The number of items in train is: 26 The loss for epoch 7 0.4773519867314742
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-history.jsonl
The running loss is: 9.447976488620043 The number of items in train is: 26 The loss for epoch 8 0.3633837111007709
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-history.jsonl
The running loss is: 9.989659074693918 The number of items in train is: 26 The loss for epoch 9 0.3842176567189969 1 Data saved to: 28_May_202005_39PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_39PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 10, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 578.508301
66 2020-04-22 sub_region ... 66 578.518188
67 2020-04-23 sub_region ... 67 578.516479
68 2020-04-24 sub_region ... 68 578.510986
69 2020-04-25 sub_region ... 69 578.503906
70 2020-04-26 sub_region ... 70 578.503906
71 2020-04-27 sub_region ... 71 578.500427
72 2020-04-28 sub_region ... 72 578.506470
73 2020-04-29 sub_region ... 73 578.516052
74 2020-04-30 sub_region ... 74 578.512817
[20 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: yrbu62av
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/media/plotly/test_plot_all_21_53c27da0.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/media/plotly/test_plot_20_ae1f0e4f.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_173944-yrbu62av/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: yrbu62av INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: awwp0r4a with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 1 use_mask: True wandb: Agent Started Run: awwp0r4a
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/awwp0r4a INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmF3d3AwcjRhOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
11.178941723890603
The number of items in train is:
27
The loss for epoch 0
0.4140348786626149
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/media/graph/graph_0_summary_6d0683bd.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/media
The running loss is: 14.482956442050636 The number of items in train is: 27 The loss for epoch 1 0.5364057941500235 1 The running loss is: 12.16446726070717 The number of items in train is: 27 The loss for epoch 2 0.4505358244706359
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-summary.json
The running loss is: 7.8048704912635 The number of items in train is: 27 The loss for epoch 3 0.2890692774542037 1 The running loss is: 7.099390174262226 The number of items in train is: 27 The loss for epoch 4 0.26294037682452687 The running loss is: 6.51896614592988 The number of items in train is: 27 The loss for epoch 5 0.24144319058999558
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-summary.json
The running loss is: 6.50308578943077 The number of items in train is: 27 The loss for epoch 6 0.24085502923817667 1 The running loss is: 6.709017640678212 The number of items in train is: 27 The loss for epoch 7 0.24848213483993378 The running loss is: 6.639082251727814 The number of items in train is: 27 The loss for epoch 8 0.2458919352491783 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-summary.json
The running loss is: 5.743373372781207 The number of items in train is: 27 The loss for epoch 9 0.21271753232522989 Data saved to: 28_May_202005_40PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_40PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['awwp0r4a']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-summary.json /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 758.606506
66 2020-04-22 sub_region ... 66 838.690002
67 2020-04-23 sub_region ... 67 739.158081
68 2020-04-24 sub_region ... 68 760.182251
69 2020-04-25 sub_region ... 69 791.689392
70 2020-04-26 sub_region ... 70 821.668945
71 2020-04-27 sub_region ... 71 813.191467
72 2020-04-28 sub_region ... 72 824.506714
73 2020-04-29 sub_region ... 73 810.240173
74 2020-04-30 sub_region ... 74 760.039307
[21 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/media/plotly/test_plot_20_50b5ba08.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: awwp0r4a
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174000-awwp0r4a/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: awwp0r4a INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 2a1ypu55 with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 1 use_mask: False wandb: Agent Started Run: 2a1ypu55
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/2a1ypu55 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjJhMXlwdTU1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
11.243638909421861
The number of items in train is:
27
The loss for epoch 0
0.41643107071932817
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/media/graph/graph_0_summary_7e02471c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/media/graph
The running loss is: 15.065432488219813 The number of items in train is: 27 The loss for epoch 1 0.5579789810451783 1 The running loss is: 9.927668149583042 The number of items in train is: 27 The loss for epoch 2 0.36769141294752006
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-summary.json
The running loss is: 7.617937269620597 The number of items in train is: 27 The loss for epoch 3 0.2821458248007629 1 The running loss is: 6.723564739339054 The number of items in train is: 27 The loss for epoch 4 0.2490209162718168 The running loss is: 7.1353028768789954 The number of items in train is: 27 The loss for epoch 5 0.2642704769214443 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-summary.json
The running loss is: 5.726659771054983 The number of items in train is: 27 The loss for epoch 6 0.21209851003907346 2 The running loss is: 6.842962785856798 The number of items in train is: 27 The loss for epoch 7 0.2534430661428444 The running loss is: 6.147264284198172 The number of items in train is: 27 The loss for epoch 8 0.22767645497030267 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-summary.json
The running loss is: 6.772955868189456 The number of items in train is: 27 The loss for epoch 9 0.25085021734035023 Data saved to: 28_May_202005_40PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['2a1ypu55']
Data saved to: 28_May_202005_40PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 702.101074
66 2020-04-22 sub_region ... 66 799.573730
67 2020-04-23 sub_region ... 67 688.857788
68 2020-04-24 sub_region ... 68 695.529663
69 2020-04-25 sub_region ... 69 724.923462
70 2020-04-26 sub_region ... 70 761.772095
71 2020-04-27 sub_region ... 71 748.695679
72 2020-04-28 sub_region ... 72 803.100220
73 2020-04-29 sub_region ... 73 733.273804
74 2020-04-30 sub_region ... 74 718.350464
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/media/plotly/test_plot_20_e049cd21.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 2a1ypu55
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174017-2a1ypu55/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 2a1ypu55 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 5py2oj8v with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 2 use_mask: True wandb: Agent Started Run: 5py2oj8v
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/5py2oj8v INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-metadata.json INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjVweTJvajh2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
13.754779417067766
The number of items in train is:
27
The loss for epoch 0
0.5094362747062136
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/media/graph/graph_0_summary_fbdf70f1.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/media
The running loss is: 15.335549898445606 The number of items in train is: 27 The loss for epoch 1 0.5679833295720594 1 The running loss is: 12.792411483824253 The number of items in train is: 27 The loss for epoch 2 0.4737930179194168 The running loss is: 10.463905839249492 The number of items in train is: 27 The loss for epoch 3 0.38755206812035153
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-summary.json
1 The running loss is: 8.178016487509012 The number of items in train is: 27 The loss for epoch 4 0.3028894995373708 The running loss is: 7.196833549998701 The number of items in train is: 27 The loss for epoch 5 0.2665493907406926
INFO:wandb.wandb_agent:Running runs: ['5py2oj8v']
1 The running loss is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-history.jsonl
7.724783644080162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-summary.json
The number of items in train is: 27 The loss for epoch 6 0.2861030979288949 2 The running loss is: 7.328360717743635 The number of items in train is: 27 The loss for epoch 7 0.27142076732383835 The running loss is: 8.112628399394453 The number of items in train is: 27 The loss for epoch 8 0.30046771849609083 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-summary.json
The running loss is: 6.612887730821967 The number of items in train is: 27 The loss for epoch 9 0.244921767808221 Data saved to: 28_May_202005_40PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_40PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 785.347290
66 2020-04-22 sub_region ... 66 848.155396
67 2020-04-23 sub_region ... 67 778.021973
68 2020-04-24 sub_region ... 68 802.694946
69 2020-04-25 sub_region ... 69 814.782837
70 2020-04-26 sub_region ... 70 827.907227
71 2020-04-27 sub_region ... 71 817.994751
72 2020-04-28 sub_region ... 72 834.097107
73 2020-04-29 sub_region ... 73 812.438049
74 2020-04-30 sub_region ... 74 797.486328
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/media/plotly/test_plot_20_e9ff783e.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 5py2oj8v
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174033-5py2oj8v/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 5py2oj8v INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: f0fvvzze with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 2 use_mask: False wandb: Agent Started Run: f0fvvzze
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/f0fvvzze INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmYwZnZ2enplOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
13.334896691143513
The number of items in train is:
27
The loss for epoch 0
0.4938850626349449
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/media/graph/graph_0_summary_23078bda.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/media
The running loss is: 15.006994970142841 The number of items in train is: 27 The loss for epoch 1 0.5558146285238089 1 The running loss is: 11.964153863489628 The number of items in train is: 27 The loss for epoch 2 0.44311680975887513
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-history.jsonl
The running loss is: 9.46366179548204 The number of items in train is: 27 The loss for epoch 3 0.3505059924252607 1 The running loss is: 7.935978587716818 The number of items in train is: 27 The loss for epoch 4 0.29392513287840066 The running loss is: 7.160205790773034 The number of items in train is: 27 The loss for epoch 5 0.26519280706566795 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-history.jsonl
The running loss is: 7.432532018050551 The number of items in train is: 27 The loss for epoch 6 0.2752789636315019 2 The running loss is: 6.930170862004161 The number of items in train is: 27 The loss for epoch 7 0.256672994889043 The running loss is: 7.583688434679061 The number of items in train is: 27 The loss for epoch 8 0.2808773494325578 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-history.jsonl
The running loss is: 6.852206656709313 The number of items in train is: 27 The loss for epoch 9 0.2537854317299746 Data saved to: 28_May_202005_40PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_40PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['f0fvvzze'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 792.838745
66 2020-04-22 sub_region ... 66 839.469238
67 2020-04-23 sub_region ... 67 793.361755
68 2020-04-24 sub_region ... 68 806.429504
69 2020-04-25 sub_region ... 69 816.609619
70 2020-04-26 sub_region ... 70 828.707764
71 2020-04-27 sub_region ... 71 823.823608
72 2020-04-28 sub_region ... 72 829.279419
73 2020-04-29 sub_region ... 73 820.153503
74 2020-04-30 sub_region ... 74 806.989746
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/media/plotly/test_plot_20_4bc95361.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: f0fvvzze
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174050-f0fvvzze/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: f0fvvzze INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: yrjbk6sj with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 3 use_mask: True wandb: Agent Started Run: yrjbk6sj
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/yrjbk6sj INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnlyamJrNnNqOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
18.989742126315832
The number of items in train is:
26
The loss for epoch 0
0.7303746971659936
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/media/graph/graph_0_summary_55f1f535.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/media
The running loss is: 16.38419210910797 The number of items in train is: 26 The loss for epoch 1 0.6301612349656912 1 The running loss is: 15.279926925897598 The number of items in train is: 26 The loss for epoch 2 0.5876894971499076
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-history.jsonl
The running loss is: 10.945733822882175 The number of items in train is: 26 The loss for epoch 3 0.4209897624185452 1 The running loss is: 5.606289468705654 The number of items in train is: 26 The loss for epoch 4 0.21562651802714056 2 The running loss is: 4.256948810070753 The number of items in train is: 26 The loss for epoch 5 0.16372880038733667
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-history.jsonl
The running loss is: 4.631639575585723 The number of items in train is: 26 The loss for epoch 6 0.17813998367637396 1 The running loss is: 6.9152334509417415 The number of items in train is: 26 The loss for epoch 7 0.26597051734391314 The running loss is: 3.967709585092962 The number of items in train is: 26 The loss for epoch 8 0.15260421481126776
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-history.jsonl
The running loss is: 3.5715523255057633 The number of items in train is: 26 The loss for epoch 9 0.13736739713483706 Data saved to: 28_May_202005_41PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_41PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['yrjbk6sj']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 814.696899
66 2020-04-22 sub_region ... 66 825.249207
67 2020-04-23 sub_region ... 67 843.944824
68 2020-04-24 sub_region ... 68 816.804077
69 2020-04-25 sub_region ... 69 796.358887
70 2020-04-26 sub_region ... 70 799.779358
71 2020-04-27 sub_region ... 71 801.697021
72 2020-04-28 sub_region ... 72 800.930298
73 2020-04-29 sub_region ... 73 815.445740
74 2020-04-30 sub_region ... 74 822.305115
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-history.jsonl
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/media/plotly/test_plot_20_d82f34c9.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: yrjbk6sj
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174106-yrjbk6sj/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: yrjbk6sj INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: xcgr1z21 with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 3 use_mask: False wandb: Agent Started Run: xcgr1z21
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/xcgr1z21 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnhjZ3IxejIxOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
18.905425552278757
The number of items in train is:
26
The loss for epoch 0
0.7271317520107214
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/media/graph/graph_0_summary_b755edb4.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/media/graph
The running loss is: 16.324342384934425 The number of items in train is: 26 The loss for epoch 1 0.6278593224974779 1 The running loss is: 15.281524695456028 The number of items in train is: 26 The loss for epoch 2 0.5877509498252318
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-summary.json
The running loss is: 10.686566263437271 The number of items in train is: 26 The loss for epoch 3 0.41102177936297196 1 The running loss is: 6.059355471283197 The number of items in train is: 26 The loss for epoch 4 0.2330521335108922 The running loss is: 4.697019984945655 The number of items in train is: 26 The loss for epoch 5 0.1806546148056021
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-summary.json
The running loss is: 6.977002476342022 The number of items in train is: 26 The loss for epoch 6 0.26834624909007776 1 The running loss is: 5.771942357998341 The number of items in train is: 26 The loss for epoch 7 0.2219977829999362 The running loss is: 4.5645317947492 The number of items in train is: 26 The loss for epoch 8 0.17555891518266156
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-summary.json
The running loss is: 5.686972144991159 The number of items in train is: 26 The loss for epoch 9 0.21872969788427538 1 Data saved to: 28_May_202005_41PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_41PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['xcgr1z21'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 736.851746
66 2020-04-22 sub_region ... 66 772.893066
67 2020-04-23 sub_region ... 67 762.027588
68 2020-04-24 sub_region ... 68 752.332520
69 2020-04-25 sub_region ... 69 736.720947
70 2020-04-26 sub_region ... 70 741.464417
71 2020-04-27 sub_region ... 71 733.462769
72 2020-04-28 sub_region ... 72 760.639526
73 2020-04-29 sub_region ... 73 746.799805
74 2020-04-30 sub_region ... 74 770.807861
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/media/plotly/test_plot_20_556d85cf.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: xcgr1z21
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174122-xcgr1z21/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: xcgr1z21 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 8a3n97xz with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 4 use_mask: True wandb: Agent Started Run: 8a3n97xz
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/8a3n97xz INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjhhM245N3h6OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
16.948153391480446
The number of items in train is:
26
The loss for epoch 0
0.6518520535184786
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/media/graph/graph_0_summary_8afade4d.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/media
The running loss is: 13.85379159078002 The number of items in train is: 26 The loss for epoch 1 0.5328381381069238 The running loss is: 9.64835274219513 The number of items in train is: 26 The loss for epoch 2 0.37109049008442807 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-history.jsonl
The running loss is: 8.4732926171273 The number of items in train is: 26 The loss for epoch 3 0.3258958698895115 The running loss is: 5.7244476936757565 The number of items in train is: 26 The loss for epoch 4 0.22017106514137524 The running loss is: 6.5513798370957375 The number of items in train is: 26 The loss for epoch 5 0.25197614758060527 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-history.jsonl
The running loss is: 8.37942417897284 The number of items in train is: 26 The loss for epoch 6 0.3222855453451092 The running loss is: 4.463485406711698 The number of items in train is: 26 The loss for epoch 7 0.1716725156427576 1 The running loss is: 4.722438246943057 The number of items in train is: 26 The loss for epoch 8 0.18163224026704064
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-history.jsonl
The running loss is: 4.340588949620724 The number of items in train is: 26 The loss for epoch 9 0.1669457288315663 1 Data saved to: 28_May_202005_41PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_41PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['8a3n97xz'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 851.006592
66 2020-04-22 sub_region ... 66 861.678711
67 2020-04-23 sub_region ... 67 870.574524
68 2020-04-24 sub_region ... 68 854.424133
69 2020-04-25 sub_region ... 69 835.877502
70 2020-04-26 sub_region ... 70 840.583374
71 2020-04-27 sub_region ... 71 839.167969
72 2020-04-28 sub_region ... 72 850.335999
73 2020-04-29 sub_region ... 73 848.938354
74 2020-04-30 sub_region ... 74 859.935547
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/media/plotly/test_plot_20_0f4d1d99.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 8a3n97xz
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174139-8a3n97xz/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 8a3n97xz INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 13w5kfa8 with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 4 use_mask: False wandb: Agent Started Run: 13w5kfa8
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/13w5kfa8 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjEzdzVrZmE4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
16.634709760546684
The number of items in train is:
26
The loss for epoch 0
0.6397965292517955
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/media/graph/graph_0_summary_96db89bb.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/media
The running loss is: 14.594469293951988 The number of items in train is: 26 The loss for epoch 1 0.5613257420750765 The running loss is: 9.017393708229065 The number of items in train is: 26 The loss for epoch 2 0.3468228349318871 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-summary.json
The running loss is: 7.2338755913078785 The number of items in train is: 26 The loss for epoch 3 0.27822598428107226 The running loss is: 4.978128891438246 The number of items in train is: 26 The loss for epoch 4 0.19146649582454792 1 The running loss is: 5.498803533613682 The number of items in train is: 26 The loss for epoch 5 0.21149244360052621 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-summary.json
The running loss is: 7.14033218100667 The number of items in train is: 26 The loss for epoch 6 0.27462816080794883 3 Stopping model now Data saved to: 28_May_202005_41PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_41PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 736.526611
66 2020-04-22 sub_region ... 66 758.064575
67 2020-04-23 sub_region ... 67 751.773193
68 2020-04-24 sub_region ... 68 746.766174
69 2020-04-25 sub_region ... 69 738.920166
70 2020-04-26 sub_region ... 70 743.997314
71 2020-04-27 sub_region ... 71 741.321045
72 2020-04-28 sub_region ... 72 750.091064
73 2020-04-29 sub_region ... 73 743.901367
74 2020-04-30 sub_region ... 74 754.618958
[21 rows x 28 columns]
INFO:wandb.wandb_agent:Running runs: ['13w5kfa8']
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/media/plotly/test_plot_14_251cab71.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 13w5kfa8
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/media/plotly/test_plot_all_15_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174155-13w5kfa8/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 13w5kfa8 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: xmks4586 with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 5 use_mask: True wandb: Agent Started Run: xmks4586
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/xmks4586 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/wandb-metadata.json INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnhta3M0NTg2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
14.528097435832024
The number of items in train is:
25
The loss for epoch 0
0.581123897433281
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/config.yaml
The running loss is: 14.321749061346054 The number of items in train is: 25 The loss for epoch 1 0.5728699624538421
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/media/graph/graph_0_summary_a2ecbda0.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/media
The running loss is: 8.53690830618143 The number of items in train is: 25 The loss for epoch 2 0.3414763322472572 1 The running loss is: 3.877267986536026 The number of items in train is: 25 The loss for epoch 3 0.15509071946144104
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/wandb-history.jsonl
The running loss is: 3.9804319692775607 The number of items in train is: 25 The loss for epoch 4 0.15921727877110242 1 The running loss is: 4.088011069223285 The number of items in train is: 25 The loss for epoch 5 0.16352044276893138
INFO:wandb.wandb_agent:Running runs: ['xmks4586']
The running loss is: 3.642200954724103 The number of items in train is: 25 The loss for epoch 6 0.14568803818896414 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/wandb-history.jsonl
The running loss is: 4.692892727442086 The number of items in train is: 25 The loss for epoch 7 0.18771570909768343 2 The running loss is: 3.2474756240844727 The number of items in train is: 25 The loss for epoch 8 0.1298990249633789 The running loss is: 3.19531544810161 The number of items in train is: 25 The loss for epoch 9 0.1278126179240644 Data saved to: 28_May_202005_42PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/wandb-history.jsonl
Data saved to: 28_May_202005_42PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 711.413086
66 2020-04-22 sub_region ... 66 714.489502
67 2020-04-23 sub_region ... 67 749.420166
68 2020-04-24 sub_region ... 68 730.427063
69 2020-04-25 sub_region ... 69 707.012146
70 2020-04-26 sub_region ... 70 700.846252
71 2020-04-27 sub_region ... 71 701.634644
72 2020-04-28 sub_region ... 72 702.439697
73 2020-04-29 sub_region ... 73 710.740356
74 2020-04-30 sub_region ... 74 736.382812
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: xmks4586
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/media/plotly/test_plot_20_b4b65440.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174208-xmks4586/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: xmks4586 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: yq5x8xdo with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 1 out_seq_length: 5 use_mask: False wandb: Agent Started Run: yq5x8xdo
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/yq5x8xdo INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnlxNXg4eGRvOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/media/graph/graph_0_summary_ec8df8ad.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/media
The running loss is: 14.583468616008759 The number of items in train is: 25 The loss for epoch 0 0.5833387446403503 The running loss is: 11.234198205173016 The number of items in train is: 25 The loss for epoch 1 0.4493679282069206 1 The running loss is: 6.8342317044734955 The number of items in train is: 25
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-summary.json
The loss for epoch 2 0.2733692681789398 2 The running loss is: 4.914771635085344 The number of items in train is: 25 The loss for epoch 3 0.19659086540341378 The running loss is: 6.097113943658769 The number of items in train is: 25 The loss for epoch 4 0.24388455774635076 1 The running loss is: 5.855831284075975 The number of items in train is: 25 The loss for epoch 5
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-history.jsonl
0.23423325136303902
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-summary.json
2 The running loss is: 4.823610462248325 The number of items in train is: 25 The loss for epoch 6 0.19294441848993302 The running loss is: 5.116255223751068 The number of items in train is: 25 The loss for epoch 7 0.2046502089500427 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-summary.json
The running loss is: 3.9354080138728023 The number of items in train is: 25 The loss for epoch 8 0.1574163205549121 2 The running loss is: 3.635712749324739 The number of items in train is: 25 The loss for epoch 9 0.14542850997298956 3 Stopping model now Data saved to: 28_May_202005_42PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['yq5x8xdo']
Data saved to: 28_May_202005_42PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 742.793396
66 2020-04-22 sub_region ... 66 736.652344
67 2020-04-23 sub_region ... 67 754.830688
68 2020-04-24 sub_region ... 68 749.173950
69 2020-04-25 sub_region ... 69 738.829346
70 2020-04-26 sub_region ... 70 734.617004
71 2020-04-27 sub_region ... 71 732.794556
72 2020-04-28 sub_region ... 72 736.974976
73 2020-04-29 sub_region ... 73 737.531616
74 2020-04-30 sub_region ... 74 753.737549
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/media/plotly/test_plot_20_71421788.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: yq5x8xdo
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174225-yq5x8xdo/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: yq5x8xdo INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 76vrqty5 with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 1 use_mask: True wandb: Agent Started Run: 76vrqty5
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/76vrqty5 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/wandb-metadata.json INFO:wandb.wandb_agent:Running runs: ['76vrqty5'] INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/config.yaml
The running loss is: 12.349453534625354 The number of items in train is: 27 The loss for epoch 0 0.4573871679490872
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/media/graph/graph_0_summary_929ab4e3.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/media/graph
The running loss is: 30.535183822968975 The number of items in train is: 27 The loss for epoch 1 1.130932734184036
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/wandb-history.jsonl
The running loss is: 18.304419979453087 The number of items in train is: 27 The loss for epoch 2 0.6779414807204847 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/wandb-history.jsonl
The running loss is: 10.870923589682207 The number of items in train is: 27 The loss for epoch 3 0.4026267996178595 2 The running loss is: 9.898105231113732 The number of items in train is: 27 The loss for epoch 4 0.36659649004124933 3 Stopping model now Data saved to: 28_May_202005_42PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/wandb-history.jsonl
Data saved to: 28_May_202005_42PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 502.529541
66 2020-04-22 sub_region ... 66 492.594849
67 2020-04-23 sub_region ... 67 496.524628
68 2020-04-24 sub_region ... 68 495.569824
69 2020-04-25 sub_region ... 69 495.434570
70 2020-04-26 sub_region ... 70 493.173859
71 2020-04-27 sub_region ... 71 493.890808
72 2020-04-28 sub_region ... 72 491.183136
73 2020-04-29 sub_region ... 73 497.434570
74 2020-04-30 sub_region ... 74 489.901245
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 76vrqty5
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/media/plotly/test_plot_all_11_6655637c.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/media/plotly/test_plot_10_c557f1bd.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174241-76vrqty5/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 76vrqty5 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: vso6t8md with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 1 use_mask: False wandb: Agent Started Run: vso6t8md
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/vso6t8md INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-metadata.json INFO:wandb.wandb_agent:Running runs: ['vso6t8md'] INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/media/graph/graph_0_summary_0ea44e09.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/media
The running loss is: 12.00125639885664 The number of items in train is: 27 The loss for epoch 0 0.4444909777354311 The running loss is: 27.564760353183374 The number of items in train is: 27 The loss for epoch 1 1.0209170501179028
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-summary.json
The running loss is: 19.1493139838567 The number of items in train is: 27 The loss for epoch 2 0.7092338512539519 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-summary.json
The running loss is: 9.482985021779314 The number of items in train is: 27 The loss for epoch 3 0.3512216674733079 The running loss is: 8.379846808500588 The number of items in train is: 27 The loss for epoch 4 0.31036469661113286 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-summary.json
The running loss is: 6.564052293615532 The number of items in train is: 27 The loss for epoch 5 0.24311304791168636 The running loss is: 7.043116731874761 The number of items in train is: 27 The loss for epoch 6 0.2608561752546208
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-summary.json
The running loss is: 6.5104550529504195 The number of items in train is: 27 The loss for epoch 7 0.2411279649240896 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-summary.json
The running loss is: 7.045392430853099 The number of items in train is: 27 The loss for epoch 8 0.2609404604019666 The running loss is: 6.926817309111357 The number of items in train is: 27 The loss for epoch 9 0.25654878922634655 1 Data saved to: 28_May_202005_43PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-metadata.json
Data saved to: 28_May_202005_43PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 778.043945
66 2020-04-22 sub_region ... 66 783.455933
67 2020-04-23 sub_region ... 67 766.487671
68 2020-04-24 sub_region ... 68 773.625916
69 2020-04-25 sub_region ... 69 782.676147
70 2020-04-26 sub_region ... 70 784.909546
71 2020-04-27 sub_region ... 71 784.748901
72 2020-04-28 sub_region ... 72 778.614258
73 2020-04-29 sub_region ... 73 782.117676
74 2020-04-30 sub_region ... 74 761.960938
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/media/plotly/test_plot_20_647b6c30.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: vso6t8md
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174305-vso6t8md/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: vso6t8md INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 9dwjgh02 with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 2 use_mask: True wandb: Agent Started Run: 9dwjgh02
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/9dwjgh02 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-metadata.json INFO:wandb.wandb_agent:Running runs: ['9dwjgh02'] INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/config.yaml
The running loss is: 15.362042212858796
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/requirements.txt
The number of items in train is:
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-history.jsonl
27
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/media/graph/graph_0_summary_3bf2505d.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-summary.json
The loss for epoch 0
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/media
0.5689645264021777
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/media/graph
The running loss is: 18.604257106781006 The number of items in train is: 27 The loss for epoch 1 0.6890465595104076 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-summary.json
The running loss is: 19.387704892084002 The number of items in train is: 27 The loss for epoch 2 0.7180631441512594 The running loss is: 17.115502156317234
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-history.jsonl
The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-summary.json
27 The loss for epoch 3 0.6339074872710087
wandb: Network error resolved after 0:00:12.859720, resuming normal operation.
The running loss is: 9.399747041985393 The number of items in train is: 27 The loss for epoch 4 0.3481387793327923 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-summary.json
The running loss is: 8.062234118580818 The number of items in train is: 27 The loss for epoch 5 0.2986012636511414 The running loss is: 7.508644045330584 The number of items in train is: 27 The loss for epoch 6 0.2780979276048364
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-summary.json
The running loss is: 7.624856340698898 The number of items in train is: 27 The loss for epoch 7 0.2824020866925518 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-summary.json
The running loss is: 7.4580579325556755 The number of items in train is: 27 The loss for epoch 8 0.27622436787243243 The running loss is: 6.958448266610503 The number of items in train is: 27 The loss for epoch 9 0.25772030617075936 Data saved to: 28_May_202005_43PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-metadata.json
Data saved to: 28_May_202005_43PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 772.555298
66 2020-04-22 sub_region ... 66 766.406433
67 2020-04-23 sub_region ... 67 773.479492
68 2020-04-24 sub_region ... 68 770.044922
69 2020-04-25 sub_region ... 69 772.980347
70 2020-04-26 sub_region ... 70 772.595215
71 2020-04-27 sub_region ... 71 772.935425
72 2020-04-28 sub_region ... 72 770.757202
73 2020-04-29 sub_region ... 73 772.154602
74 2020-04-30 sub_region ... 74 771.334900
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/media/plotly/test_plot_20_cf3ed598.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 9dwjgh02
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174334-9dwjgh02/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 9dwjgh02 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: xnsxnzq6 with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 2 use_mask: False wandb: Agent Started Run: xnsxnzq6
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/xnsxnzq6 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnhuc3huenE2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/media/graph/graph_0_summary_7969e239.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/media
The running loss is: 15.227613762952387 The number of items in train is: 27 The loss for epoch 0 0.5639856949241625 The running loss is: 17.898133425042033 The number of items in train is: 27 The loss for epoch 1 0.6628938305571124 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-summary.json
The running loss is: 15.654380347579718 The number of items in train is: 27 The loss for epoch 2 0.5797918647251747
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-summary.json
The running loss is: 9.89348728954792 The number of items in train is: 27 The loss for epoch 3 0.3664254551684415 1 The running loss is: 8.431944162817672 The number of items in train is: 27 The loss for epoch 4 0.31229422825250636
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-summary.json
The running loss is: 8.609166414942592 The number of items in train is: 27 The loss for epoch 5 0.31885801536824415 1
INFO:wandb.wandb_agent:Running runs: ['xnsxnzq6']
The running loss is: 9.016866168007255 The number of items in train is: 27 The loss for epoch 6 0.3339580062224909 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-summary.json
The running loss is: 8.0558643322438 The number of items in train is: 27 The loss for epoch 7 0.29836534563865924
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-summary.json
The running loss is: 8.685117572546005 The number of items in train is: 27 The loss for epoch 8 0.3216710212054076 1 The running loss is: 6.745234100148082 The number of items in train is: 27 The loss for epoch 9 0.2498234851906697 Data saved to: 28_May_202005_44PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-summary.json
Data saved to: 28_May_202005_44PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 783.742432
66 2020-04-22 sub_region ... 66 782.695801
67 2020-04-23 sub_region ... 67 783.955322
68 2020-04-24 sub_region ... 68 783.325439
69 2020-04-25 sub_region ... 69 783.187744
70 2020-04-26 sub_region ... 70 783.111938
71 2020-04-27 sub_region ... 71 783.345886
72 2020-04-28 sub_region ... 72 782.962280
73 2020-04-29 sub_region ... 73 782.751831
74 2020-04-30 sub_region ... 74 783.127808
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: xnsxnzq6
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/media/plotly/test_plot_20_86f33ef2.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174400-xnsxnzq6/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: xnsxnzq6 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: tr5rhjwy with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 3 use_mask: True wandb: Agent Started Run: tr5rhjwy
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/tr5rhjwy INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnRyNXJoand5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/media/graph/graph_0_summary_ff3d77af.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/media
The running loss is: 21.012039236724377 The number of items in train is: 26 The loss for epoch 0 0.8081553552586299 The running loss is: 16.833910040557384 The number of items in train is: 26 The loss for epoch 1 0.6474580784829763 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-summary.json
The running loss is: 16.23337048664689 The number of items in train is: 26 The loss for epoch 2 0.6243604033325727
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-summary.json
The running loss is: 15.538650223985314 The number of items in train is: 26 The loss for epoch 3 0.5976403932302043 1 The running loss is: 15.044064596295357 The number of items in train is: 26 The loss for epoch 4 0.5786178690882829 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-summary.json
The running loss is: 11.831363804638386 The number of items in train is: 26 The loss for epoch 5 0.4550524540245533
INFO:wandb.wandb_agent:Running runs: ['tr5rhjwy']
The running loss is: 7.703582746908069 The number of items in train is: 26 The loss for epoch 6 0.2962916441118488 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-summary.json
The running loss is: 8.853043738752604 The number of items in train is: 26 The loss for epoch 7 0.3405016822597155 2 The running loss is: 5.49346605874598 The number of items in train is: 26 The loss for epoch 8
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-history.jsonl
0.21128715610561463
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-summary.json
The running loss is: 6.049623145721853 The number of items in train is: 26 The loss for epoch 9 0.23267781329699433 Data saved to: 28_May_202005_44PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-summary.json
Data saved to: 28_May_202005_44PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 793.348389
66 2020-04-22 sub_region ... 66 793.042908
67 2020-04-23 sub_region ... 67 790.546631
68 2020-04-24 sub_region ... 68 791.214233
69 2020-04-25 sub_region ... 69 792.006836
70 2020-04-26 sub_region ... 70 792.608032
71 2020-04-27 sub_region ... 71 792.963989
72 2020-04-28 sub_region ... 72 789.634521
73 2020-04-29 sub_region ... 73 793.287476
74 2020-04-30 sub_region ... 74 785.882202
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: tr5rhjwy
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/media/plotly/test_plot_20_ef160644.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174416-tr5rhjwy/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: tr5rhjwy INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 98gbz6d6 with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 3 use_mask: False wandb: Agent Started Run: 98gbz6d6
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/98gbz6d6 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjk4Z2J6NmQ2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/media/graph/graph_0_summary_6698a735.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/media
The running loss is: 20.714700505137444 The number of items in train is: 26 The loss for epoch 0 0.7967192501975939 The running loss is: 16.891354955732822 The number of items in train is: 26 The loss for epoch 1 0.6496674982974162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-summary.json
The running loss is: 16.139072712510824 The number of items in train is: 26 The loss for epoch 2 0.620733565865801 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-summary.json
The running loss is: 15.367559241130948 The number of items in train is: 26 The loss for epoch 3 0.5910599708127288 2 The running loss is: 14.148938469588757 The number of items in train is: 26 The loss for epoch 4 0.5441899411380291 3 Stopping model now Data saved to: 28_May_202005_44PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-summary.json
Data saved to: 28_May_202005_44PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.wandb_agent:Running runs: ['98gbz6d6']
CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 440.134735
66 2020-04-22 sub_region ... 66 439.800903
67 2020-04-23 sub_region ... 67 443.283325
68 2020-04-24 sub_region ... 68 442.842438
69 2020-04-25 sub_region ... 69 440.653381
70 2020-04-26 sub_region ... 70 441.242767
71 2020-04-27 sub_region ... 71 440.577118
72 2020-04-28 sub_region ... 72 442.682800
73 2020-04-29 sub_region ... 73 441.382904
74 2020-04-30 sub_region ... 74 445.453522
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/media/plotly/test_plot_10_0d49373a.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/media/plotly
wandb: Agent Finished Run: 98gbz6d6
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/media/plotly/test_plot_all_11_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174434-98gbz6d6/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 98gbz6d6 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: e13nqx3k with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 4 use_mask: True wandb: Agent Started Run: e13nqx3k
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/e13nqx3k INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmUxM25xeDNrOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/config.yaml
The running loss is: 23.482627481222153 The number of items in train is: 26 The loss for epoch 0 0.9031779800470059
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/media/graph/graph_0_summary_bcd87636.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/media/graph
The running loss is: 20.23513476550579 The number of items in train is: 26 The loss for epoch 1 0.778274414057915 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-summary.json
The running loss is: 21.649361565709114 The number of items in train is: 26 The loss for epoch 2 0.8326677525272737
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-summary.json
The running loss is: 15.995433270931244 The number of items in train is: 26 The loss for epoch 3 0.615208971958894 1 The running loss is: 12.781472019851208 The number of items in train is: 26 The loss for epoch 4 0.4915950776865849 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-summary.json
The running loss is: 8.850842893123627 The number of items in train is: 26 The loss for epoch 5 0.3404170343509087
INFO:wandb.wandb_agent:Running runs: ['e13nqx3k']
The running loss is: 6.8902468085289 The number of items in train is: 26 The loss for epoch 6 0.26500949263572693 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-summary.json
The running loss is: 8.579151637852192 The number of items in train is: 26 The loss for epoch 7 0.32996737068662274 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-summary.json
The running loss is: 6.014725340530276 The number of items in train is: 26 The loss for epoch 8 0.23133559002039525 The running loss is: 4.920796798542142 The number of items in train is: 26 The loss for epoch 9 0.18926141532854393 1 Data saved to: 28_May_202005_44PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-summary.json
Data saved to: 28_May_202005_44PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 784.844604
66 2020-04-22 sub_region ... 66 785.061829
67 2020-04-23 sub_region ... 67 785.951660
68 2020-04-24 sub_region ... 68 785.422729
69 2020-04-25 sub_region ... 69 784.747925
70 2020-04-26 sub_region ... 70 784.533691
71 2020-04-27 sub_region ... 71 784.408447
72 2020-04-28 sub_region ... 72 785.171936
73 2020-04-29 sub_region ... 73 784.649170
74 2020-04-30 sub_region ... 74 785.943665
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: e13nqx3k
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/media/plotly/test_plot_20_bf33d510.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174445-e13nqx3k/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: e13nqx3k INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: i222q13p with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 4 use_mask: False wandb: Agent Started Run: i222q13p
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/i222q13p INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-metadata.json INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmkyMjJxMTNwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/media/graph/graph_0_summary_55e43b9c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/media
The running loss is: 23.31422958523035 The number of items in train is: 26 The loss for epoch 0 0.896701137893475
INFO:wandb.wandb_agent:Running runs: ['i222q13p'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-history.jsonl
The running loss is: 19.628748193383217 The number of items in train is: 26 The loss for epoch 1 0.7549518535916622 1 The running loss is: 22.03840383887291 The number of items in train is: 26 The loss for epoch 2 0.8476309168797272
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-history.jsonl
The running loss is: 15.539160393178463 The number of items in train is: 26 The loss for epoch 3 0.5976600151222485 1 The running loss is: 12.818177118897438 The number of items in train is: 26 The loss for epoch 4 0.4930068122652861
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-history.jsonl
The running loss is: 10.63734956085682 The number of items in train is: 26 The loss for epoch 5 0.4091288292637238 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-history.jsonl
The running loss is: 6.949803344905376 The number of items in train is: 26 The loss for epoch 6 0.2673001286502068 The running loss is: 7.64347056671977 The number of items in train is: 26 The loss for epoch 7 0.29397963718152964 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-history.jsonl
The running loss is: 11.507175385951996 The number of items in train is: 26 The loss for epoch 8 0.4425836686904614 The running loss is: 5.667488571256399 The number of items in train is: 26 The loss for epoch 9 0.21798032966370767
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-summary.json
Data saved to:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-history.jsonl
28_May_202005_45PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_45PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 749.537842
66 2020-04-22 sub_region ... 66 753.229736
67 2020-04-23 sub_region ... 67 753.515015
68 2020-04-24 sub_region ... 68 753.205444
69 2020-04-25 sub_region ... 69 751.122925
70 2020-04-26 sub_region ... 70 750.473755
71 2020-04-27 sub_region ... 71 748.162842
72 2020-04-28 sub_region ... 72 752.245361
73 2020-04-29 sub_region ... 73 750.474243
74 2020-04-30 sub_region ... 74 754.320679
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: i222q13p
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/media/plotly/test_plot_20_67e12e5e.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174501-i222q13p/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: i222q13p INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: f6en72ry with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 5 use_mask: True wandb: Agent Started Run: f6en72ry
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/f6en72ry INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmY2ZW43MnJ5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/media/graph/graph_0_summary_9ff0aff4.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/media/graph
The running loss is: 17.830005079507828 The number of items in train is: 25 The loss for epoch 0 0.7132002031803131 The running loss is: 16.9639795422554 The number of items in train is: 25 The loss for epoch 1 0.6785591816902161 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-summary.json
The running loss is: 19.417406290769577 The number of items in train is: 25 The loss for epoch 2 0.776696251630783 The running loss is: 13.737709395587444 The number of items in train is: 25 The loss for epoch 3 0.5495083758234978 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-summary.json
The running loss is: 13.103170596063137 The number of items in train is: 25 The loss for epoch 4 0.5241268238425255 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-summary.json
The running loss is: 9.996175795793533 The number of items in train is: 25 The loss for epoch 5 0.3998470318317413 3 Stopping model now Data saved to: 28_May_202005_45PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_45PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['f6en72ry'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 417.858063
66 2020-04-22 sub_region ... 66 419.026245
67 2020-04-23 sub_region ... 67 420.017090
68 2020-04-24 sub_region ... 68 420.572571
69 2020-04-25 sub_region ... 69 419.899597
70 2020-04-26 sub_region ... 70 419.641113
71 2020-04-27 sub_region ... 71 419.378143
72 2020-04-28 sub_region ... 72 419.695251
73 2020-04-29 sub_region ... 73 419.175232
74 2020-04-30 sub_region ... 74 421.516815
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/media/plotly/test_plot_12_d17e53a2.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: f6en72ry
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/media/plotly/test_plot_all_13_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174523-f6en72ry/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: f6en72ry INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 3et7fdzd with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 2 out_seq_length: 5 use_mask: False wandb: Agent Started Run: 3et7fdzd
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/3et7fdzd INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjNldDdmZHpkOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/media/graph/graph_0_summary_29229bcb.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/media
The running loss is: 17.741086795926094 The number of items in train is: 25 The loss for epoch 0 0.7096434718370438 The running loss is: 16.730679839849472 The number of items in train is: 25 The loss for epoch 1 0.6692271935939789 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-summary.json
The running loss is: 19.939407482743263 The number of items in train is: 25 The loss for epoch 2 0.7975762993097305 The running loss is: 13.255669929087162 The number of items in train is: 25 The loss for epoch 3 0.5302267971634865
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-summary.json
1 The running loss is: 11.027616888284683 The number of items in train is: 25 The loss for epoch 4 0.4411046755313873 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-summary.json
The running loss is: 6.718288138508797 The number of items in train is: 25 The loss for epoch 5 0.26873152554035185 The running loss is: 7.313029149547219 The number of items in train is: 25 The loss for epoch 6 0.29252116598188876 1
INFO:wandb.wandb_agent:Running runs: ['3et7fdzd'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-summary.json
The running loss is: 12.773601979017258 The number of items in train is: 25 The loss for epoch 7 0.5109440791606903 The running loss is: 6.80379575304687 The number of items in train is: 25 The loss for epoch 8 0.2721518301218748 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-summary.json
The running loss is: 4.739170776680112 The number of items in train is: 25 The loss for epoch 9 0.18956683106720448 2 Data saved to: 28_May_202005_45PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_45PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 770.853760
66 2020-04-22 sub_region ... 66 769.976440
67 2020-04-23 sub_region ... 67 771.220154
68 2020-04-24 sub_region ... 68 770.911987
69 2020-04-25 sub_region ... 69 770.517700
70 2020-04-26 sub_region ... 70 770.186768
71 2020-04-27 sub_region ... 71 770.193970
72 2020-04-28 sub_region ... 72 770.157593
73 2020-04-29 sub_region ... 73 770.232788
74 2020-04-30 sub_region ... 74 770.960083
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/media/plotly/test_plot_20_57ec849c.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 3et7fdzd
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174539-3et7fdzd/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 3et7fdzd INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: dqyud2we with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 1 use_mask: True wandb: Agent Started Run: dqyud2we
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/dqyud2we INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmRxeXVkMndlOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/media/graph/graph_0_summary_b022e570.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/media
The running loss is: 13.64892402663827 The number of items in train is: 27 The loss for epoch 0 0.5055157046903063
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-summary.json
The running loss is: 37.692751033231616 The number of items in train is: 27 The loss for epoch 1 1.3960278160456154
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-summary.json
The running loss is: 20.47567129507661 The number of items in train is: 27 The loss for epoch 2 0.7583581961139485
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-summary.json
The running loss is: 20.933722829446197 The number of items in train is: 27 The loss for epoch 3 0.7753230677572666 1
INFO:wandb.wandb_agent:Running runs: ['dqyud2we'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-summary.json
The running loss is: 20.798208009451628 The number of items in train is: 27 The loss for epoch 4 0.7703040003500603 The running loss is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-summary.json
20.184429235756397 The number of items in train is: 27 The loss for epoch 5 0.7475714531761629 1 The running loss is: 20.411993622779846 The number of items in train is: 27 The loss for epoch 6 0.755999763806661 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-summary.json
The running loss is: 19.94804622977972 The number of items in train is: 27 The loss for epoch 7 0.7388165270288786 3 Stopping model now Data saved to: 28_May_202005_46PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-summary.json
Data saved to: 28_May_202005_46PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 474.087341
66 2020-04-22 sub_region ... 66 474.107361
67 2020-04-23 sub_region ... 67 473.755005
68 2020-04-24 sub_region ... 68 474.003113
69 2020-04-25 sub_region ... 69 474.199402
70 2020-04-26 sub_region ... 70 474.221405
71 2020-04-27 sub_region ... 71 474.191833
72 2020-04-28 sub_region ... 72 473.996887
73 2020-04-29 sub_region ... 73 474.185181
74 2020-04-30 sub_region ... 74 473.697998
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: dqyud2we
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/media/plotly/test_plot_16_1f8d7214.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/media/plotly/test_plot_all_17_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174555-dqyud2we/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: dqyud2we INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: k9df4wo5 with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 1 use_mask: False wandb: Agent Started Run: k9df4wo5
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/k9df4wo5 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOms5ZGY0d281OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/media/graph/graph_0_summary_882f1ddc.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/media
The running loss is: 13.37478191871196 The number of items in train is: 27 The loss for epoch 0 0.49536229328562814
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-summary.json
The running loss is: 37.45775743201375 The number of items in train is: 27 The loss for epoch 1 1.3873243493338425
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-summary.json
The running loss is: 20.595209177583456 The number of items in train is: 27 The loss for epoch 2 0.7627855250956835
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-summary.json
The running loss is: 21.006036780774593 The number of items in train is: 27 The loss for epoch 3 0.7780013622509109
INFO:wandb.wandb_agent:Running runs: ['k9df4wo5'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-summary.json
The running loss is: 20.73020259477198 The number of items in train is: 27 The loss for epoch 4 0.7677852812878512 The running loss is: 20.24301341921091 The number of items in train is: 27 The loss for epoch 5 0.7497412377485523
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-summary.json
1 The running loss is: 20.167157787829638 The number of items in train is: 27 The loss for epoch 6 0.7469317699196162 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-summary.json
The running loss is: 18.895125970244408 The number of items in train is: 27 The loss for epoch 7 0.6998194803794225 3 Stopping model now Data saved to: 28_May_202005_46PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-summary.json
Data saved to: 28_May_202005_46PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/config.yaml /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 474.395355
66 2020-04-22 sub_region ... 66 474.474243
67 2020-04-23 sub_region ... 67 473.946503
68 2020-04-24 sub_region ... 68 474.251740
69 2020-04-25 sub_region ... 69 474.498688
70 2020-04-26 sub_region ... 70 474.531128
71 2020-04-27 sub_region ... 71 474.499969
72 2020-04-28 sub_region ... 72 474.257050
73 2020-04-29 sub_region ... 73 474.527222
74 2020-04-30 sub_region ... 74 473.818420
[21 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/media/plotly/test_plot_16_c257944a.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: k9df4wo5
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/media/plotly/test_plot_all_17_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174612-k9df4wo5/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: k9df4wo5 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: vzgo44b3 with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 2 use_mask: True wandb: Agent Started Run: vzgo44b3
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/vzgo44b3 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnZ6Z280NGIzOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/media/graph/graph_0_summary_aa1f8d8a.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/media
The running loss is: 15.137208757922053 The number of items in train is: 27 The loss for epoch 0 0.5606373614045205
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-history.jsonl
The running loss is: 30.88287841156125 The number of items in train is: 27 The loss for epoch 1 1.1438103115393057
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-history.jsonl
The running loss is: 25.371903955936432 The number of items in train is: 27 The loss for epoch 2 0.9397001465161642
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-history.jsonl
The running loss is: 20.891498629003763 The number of items in train is: 27 The loss for epoch 3 0.7737592084816208
INFO:wandb.wandb_agent:Running runs: ['vzgo44b3'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-history.jsonl
The running loss is: 20.58536821603775 The number of items in train is: 27 The loss for epoch 4 0.7624210450384352 1 The running loss is: 21.16692252457142 The number of items in train is: 27 The loss for epoch 5 0.7839600935026452 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-history.jsonl
The running loss is: 20.775313809514046 The number of items in train is: 27 The loss for epoch 6 0.7694560670190387 3 Stopping model now Data saved to: 28_May_202005_46PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-history.jsonl
Data saved to: 28_May_202005_46PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/config.yaml /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 422.063049
66 2020-04-22 sub_region ... 66 422.065277
67 2020-04-23 sub_region ... 67 422.008148
68 2020-04-24 sub_region ... 68 422.044128
69 2020-04-25 sub_region ... 69 422.075989
70 2020-04-26 sub_region ... 70 422.085114
71 2020-04-27 sub_region ... 71 422.084900
72 2020-04-28 sub_region ... 72 422.078888
73 2020-04-29 sub_region ... 73 422.060028
74 2020-04-30 sub_region ... 74 422.037537
[21 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/media/plotly/test_plot_14_7a77c25e.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: vzgo44b3
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/media/plotly/test_plot_all_15_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174628-vzgo44b3/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: vzgo44b3 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: eplaca55 with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 2 use_mask: False wandb: Agent Started Run: eplaca55
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/eplaca55 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-metadata.json INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmVwbGFjYTU1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/media/graph/graph_0_summary_e671cd5d.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/media
The running loss is: 15.071377269923687 The number of items in train is: 27 The loss for epoch 0 0.5581991581453217
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-history.jsonl
The running loss is: 31.568736284971237 The number of items in train is: 27 The loss for epoch 1 1.1692124549989347
INFO:wandb.wandb_agent:Running runs: ['eplaca55'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-history.jsonl
The running loss is: 25.198651179671288 The number of items in train is: 27 The loss for epoch 2 0.9332833770248625
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-history.jsonl
The running loss is: 21.102718703448772 The number of items in train is: 27 The loss for epoch 3 0.7815821742018064
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-history.jsonl
The running loss is: 20.57193885743618 The number of items in train is: 27 The loss for epoch 4 0.7619236613865252 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-history.jsonl
The running loss is: 21.114420972764492 The number of items in train is: 27 The loss for epoch 5 0.7820155915838701 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-history.jsonl
The running loss is: 21.067522056400776 The number of items in train is: 27 The loss for epoch 6 0.7802785946815102 3 Stopping model now Data saved to: 28_May_202005_46PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_46PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 423.483063
66 2020-04-22 sub_region ... 66 423.511169
67 2020-04-23 sub_region ... 67 423.472900
68 2020-04-24 sub_region ... 68 423.498932
69 2020-04-25 sub_region ... 69 423.520294
70 2020-04-26 sub_region ... 70 423.524902
71 2020-04-27 sub_region ... 71 423.520996
72 2020-04-28 sub_region ... 72 423.510651
73 2020-04-29 sub_region ... 73 423.502411
74 2020-04-30 sub_region ... 74 423.487152
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/media/plotly/test_plot_14_0bb7c432.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: eplaca55
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/media/plotly/test_plot_all_15_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174645-eplaca55/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: eplaca55 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: hsqsy0f0 with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 3 use_mask: True wandb: Agent Started Run: hsqsy0f0
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/hsqsy0f0 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-metadata.json INFO:wandb.wandb_agent:Running runs: ['hsqsy0f0'] INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/media/graph/graph_0_summary_51352134.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/media
The running loss is: 24.238217145204544 The number of items in train is: 26 The loss for epoch 0 0.9322391209694055
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-summary.json
The running loss is: 16.11776690930128 The number of items in train is: 26 The loss for epoch 1 0.6199141118962032 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-summary.json
The running loss is: 16.947837613523006 The number of items in train is: 26 The loss for epoch 2 0.6518399082124233 2
wandb: Network error resolved after 0:00:13.670767, resuming normal operation.
The running loss is: 17.64614214003086 The number of items in train is: 26 The loss for epoch 3 0.6786977746165715
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-summary.json
The running loss is: 16.301386021077633 The number of items in train is: 26 The loss for epoch 4 0.6269763854260628
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-summary.json
The running loss is: 16.042810605838895 The number of items in train is: 26 The loss for epoch 5 0.6170311771476498
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-summary.json
The running loss is: 15.744363751262426 The number of items in train is: 26 The loss for epoch 6 0.6055524519716318 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-metadata.json
The running loss is: 16.016306111589074 The number of items in train is: 26 The loss for epoch 7 0.6160117735226567 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-summary.json
The running loss is: 15.813427444547415 The number of items in train is: 26 The loss for epoch 8 0.6082087478672082
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-summary.json
The running loss is: 16.24589281156659 The number of items in train is: 26 The loss for epoch 9 0.6248420312140996 Data saved to: 28_May_202005_47PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_47PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9])
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-history.jsonl
Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/config.yaml /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 509.934052
66 2020-04-22 sub_region ... 66 507.839417
67 2020-04-23 sub_region ... 67 509.408234
68 2020-04-24 sub_region ... 68 509.694031
69 2020-04-25 sub_region ... 69 509.575317
70 2020-04-26 sub_region ... 70 508.830750
71 2020-04-27 sub_region ... 71 509.065674
72 2020-04-28 sub_region ... 72 507.174622
73 2020-04-29 sub_region ... 73 509.588074
74 2020-04-30 sub_region ... 74 507.856110
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/media/plotly/test_plot_20_75179c3f.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: hsqsy0f0
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174706-hsqsy0f0/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: hsqsy0f0 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: kd0jv4c2 with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 3 use_mask: False wandb: Agent Started Run: kd0jv4c2
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/kd0jv4c2 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-metadata.json INFO:wandb.wandb_agent:Running runs: ['kd0jv4c2'] INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmtkMGp2NGMyOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/media/graph/graph_0_summary_7dd40716.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/media/graph
The running loss is: 24.46081606298685 The number of items in train is: 26 The loss for epoch 0 0.9408006178071866
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-history.jsonl
The running loss is: 16.266936726868153 The number of items in train is: 26 The loss for epoch 1 0.625651412571852 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-history.jsonl
The running loss is: 16.894336327910423 The number of items in train is: 26 The loss for epoch 2 0.6497821664580932
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-history.jsonl
The running loss is: 17.256735399365425 The number of items in train is: 26 The loss for epoch 3 0.6637205922832856 The running loss is: 16.231802832335234 The number of items in train is: 26 The loss for epoch 4 0.6243001089359705
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-history.jsonl
The running loss is: 16.1003286074847 The number of items in train is: 26 The loss for epoch 5 0.6192434079801807
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-history.jsonl
The running loss is: 15.810061607509851 The number of items in train is: 26 The loss for epoch 6 0.6080792925965327 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-history.jsonl
The running loss is: 16.048187194392085 The number of items in train is: 26 The loss for epoch 7 0.6172379690150802 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-metadata.json
The running loss is: 15.809576485306025 The number of items in train is: 26 The loss for epoch 8 0.6080606340502317 3 Stopping model now Data saved to: 28_May_202005_47PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_47PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 529.040771
66 2020-04-22 sub_region ... 66 528.143555
67 2020-04-23 sub_region ... 67 528.838806
68 2020-04-24 sub_region ... 68 528.961365
69 2020-04-25 sub_region ... 69 528.801025
70 2020-04-26 sub_region ... 70 528.515015
71 2020-04-27 sub_region ... 71 528.581177
72 2020-04-28 sub_region ... 72 528.077698
73 2020-04-29 sub_region ... 73 528.772461
74 2020-04-30 sub_region ... 74 528.409302
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/media/plotly/test_plot_18_712b0439.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: kd0jv4c2
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/media/plotly/test_plot_all_19_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174739-kd0jv4c2/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: kd0jv4c2 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: b6eh4dgx with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 4 use_mask: True wandb: Agent Started Run: b6eh4dgx
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/b6eh4dgx INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmI2ZWg0ZGd4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/media/graph/graph_0_summary_11cf6ea0.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/media
The running loss is: 24.921816915273666 The number of items in train is: 26 The loss for epoch 0 0.958531419818218
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-summary.json
The running loss is: 22.33100463449955 The number of items in train is: 26 The loss for epoch 1 0.858884793634598
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-summary.json
The running loss is: 20.428308714181185 The number of items in train is: 26 The loss for epoch 2 0.7857041813146609
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-summary.json
The running loss is: 17.427796185016632 The number of items in train is: 26 The loss for epoch 3 0.6702998532698705 1
INFO:wandb.wandb_agent:Running runs: ['b6eh4dgx']
The running loss is: 17.2829820625484 The number of items in train is: 26 The loss for epoch 4 0.6647300793287846
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-summary.json
2 The running loss is: 15.887216325849295 The number of items in train is: 26 The loss for epoch 5 0.6110467817634344 3 Stopping model now Data saved to: 28_May_202005_48PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-summary.json
Data saved to: 28_May_202005_48PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 559.779724
66 2020-04-22 sub_region ... 66 560.009094
67 2020-04-23 sub_region ... 67 560.145813
68 2020-04-24 sub_region ... 68 560.165466
69 2020-04-25 sub_region ... 69 560.001709
70 2020-04-26 sub_region ... 70 560.061584
71 2020-04-27 sub_region ... 71 559.992859
72 2020-04-28 sub_region ... 72 560.228088
73 2020-04-29 sub_region ... 73 559.984314
74 2020-04-30 sub_region ... 74 560.468567
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/media/plotly/test_plot_12_f72c3f2f.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: b6eh4dgx
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/media/plotly/test_plot_all_13_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174805-b6eh4dgx/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: b6eh4dgx INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: m74n1snd with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 4 use_mask: False wandb: Agent Started Run: m74n1snd
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/m74n1snd INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm03NG4xc25kOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/media/graph/graph_0_summary_45d53722.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/media/graph
The running loss is: 25.268363758921623 The number of items in train is: 26 The loss for epoch 0 0.9718601445739086
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-history.jsonl
The running loss is: 22.410943418741226 The number of items in train is: 26 The loss for epoch 1 0.8619593622592779
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-history.jsonl
The running loss is: 20.349964793771505 The number of items in train is: 26 The loss for epoch 2 0.7826909536065964
INFO:wandb.wandb_agent:Running runs: ['m74n1snd'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-history.jsonl
The running loss is: 17.21460761502385 The number of items in train is: 26 The loss for epoch 3 0.6621002928855327 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-history.jsonl
The running loss is: 17.72370159626007 The number of items in train is: 26 The loss for epoch 4 0.6816808306253873 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-history.jsonl
The running loss is: 15.540141738951206 The number of items in train is: 26 The loss for epoch 5 0.597697759190431 3 Stopping model now Data saved to: 28_May_202005_48PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_48PM_model.pth interpolate should be below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/config.yaml
Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 559.067444
66 2020-04-22 sub_region ... 66 559.143799
67 2020-04-23 sub_region ... 67 559.442505
68 2020-04-24 sub_region ... 68 559.442566
69 2020-04-25 sub_region ... 69 559.225464
70 2020-04-26 sub_region ... 70 559.264404
71 2020-04-27 sub_region ... 71 559.198975
72 2020-04-28 sub_region ... 72 559.425171
73 2020-04-29 sub_region ... 73 559.358643
74 2020-04-30 sub_region ... 74 559.806274
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/media/plotly/test_plot_12_b8d9f787.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: m74n1snd
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/media/plotly/test_plot_all_13_6655637c.plotly.json INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/media/plotly INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174822-m74n1snd/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: m74n1snd INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: s7ektckx with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 5 use_mask: True wandb: Agent Started Run: s7ektckx
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/s7ektckx INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnM3ZWt0Y2t4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/media/graph/graph_0_summary_74300212.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/media
The running loss is: 20.4773468375206 The number of items in train is: 25 The loss for epoch 0 0.819093873500824
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-summary.json
The running loss is: 19.37137946486473 The number of items in train is: 25 The loss for epoch 1 0.7748551785945892
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-summary.json
The running loss is: 16.317528434097767 The number of items in train is: 25 The loss for epoch 2 0.6527011373639107
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-summary.json
The running loss is: 14.260967433452606 The number of items in train is: 25 The loss for epoch 3 0.5704386973381043 1
INFO:wandb.wandb_agent:Running runs: ['s7ektckx']
The running loss is: 14.17140232771635 The number of items in train is: 25 The loss for epoch 4 0.566856093108654
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-summary.json
The running loss is: 14.054600022733212 The number of items in train is: 25 The loss for epoch 5 0.5621840009093284
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-summary.json
The running loss is: 14.07607826590538 The number of items in train is: 25 The loss for epoch 6 0.5630431306362152
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-summary.json
The running loss is: 14.025826334953308 The number of items in train is: 25 The loss for epoch 7 0.5610330533981324 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-summary.json
The running loss is: 13.500305883586407 The number of items in train is: 25 The loss for epoch 8 0.5400122353434562 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-summary.json
The running loss is: 12.7346723228693 The number of items in train is: 25 The loss for epoch 9 0.509386892914772 3 Stopping model now Data saved to: 28_May_202005_48PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_48PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 384.512299
66 2020-04-22 sub_region ... 66 383.916504
67 2020-04-23 sub_region ... 67 384.334045
68 2020-04-24 sub_region ... 68 384.549500
69 2020-04-25 sub_region ... 69 384.431915
70 2020-04-26 sub_region ... 70 384.146332
71 2020-04-27 sub_region ... 71 384.198578
72 2020-04-28 sub_region ... 72 383.738586
73 2020-04-29 sub_region ... 73 384.381348
74 2020-04-30 sub_region ... 74 383.990662
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/media/plotly/test_plot_20_1987c841.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: s7ektckx
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174838-s7ektckx/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: s7ektckx INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: cda37iga with config: batch_size: 2 forecast_history: 11 lr: 0.001 number_encoder_layers: 3 out_seq_length: 5 use_mask: False wandb: Agent Started Run: cda37iga
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/cda37iga INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmNkYTM3aWdhOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/media/graph/graph_0_summary_a747039c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/media/graph
The running loss is: 20.519908547401428 The number of items in train is: 25 The loss for epoch 0 0.8207963418960571
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-history.jsonl
The running loss is: 19.348501980304718 The number of items in train is: 25 The loss for epoch 1 0.7739400792121888
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-history.jsonl
The running loss is: 16.020245015621185 The number of items in train is: 25 The loss for epoch 2 0.6408098006248474
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-history.jsonl
The running loss is: 14.293965496122837 The number of items in train is: 25 The loss for epoch 3 0.5717586198449135 1
INFO:wandb.wandb_agent:Running runs: ['cda37iga']
The running loss is: 14.129456043243408 The number of items in train is: 25 The loss for epoch 4 0.5651782417297363
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-history.jsonl
2 The running loss is: 13.794427633285522 The number of items in train is: 25 The loss for epoch 5 0.5517771053314209
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-history.jsonl
The running loss is: 13.02604092657566 The number of items in train is: 25 The loss for epoch 6 0.5210416370630264 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-history.jsonl
The running loss is: 9.591085441410542 The number of items in train is: 25 The loss for epoch 7 0.3836434176564217 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-history.jsonl
The running loss is: 7.496750168502331 The number of items in train is: 25 The loss for epoch 8 0.29987000674009323 3 Stopping model now Data saved to: 28_May_202005_49PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_49PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 388.992981
66 2020-04-22 sub_region ... 66 387.531372
67 2020-04-23 sub_region ... 67 388.493744
68 2020-04-24 sub_region ... 68 388.772583
69 2020-04-25 sub_region ... 69 388.547852
70 2020-04-26 sub_region ... 70 387.934692
71 2020-04-27 sub_region ... 71 388.074402
72 2020-04-28 sub_region ... 72 387.295563
73 2020-04-29 sub_region ... 73 388.444427
74 2020-04-30 sub_region ... 74 387.736328
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/media/plotly/test_plot_18_166901d4.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: cda37iga
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/media/plotly/test_plot_all_19_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174900-cda37iga/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: cda37iga INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: xoduj0zf with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 1 use_mask: True wandb: Agent Started Run: xoduj0zf
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/xoduj0zf INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnhvZHVqMHpmOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
11.178941723890603
The number of items in train is:
27
The loss for epoch 0
0.4140348786626149
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/media/graph/graph_0_summary_bec9b63f.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/media/graph
The running loss is: 14.482956442050636 The number of items in train is: 27 The loss for epoch 1 0.5364057941500235 1 The running loss is: 12.16446726070717 The number of items in train is: 27 The loss for epoch 2 0.4505358244706359
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-summary.json
The running loss is: 7.8048704912635 The number of items in train is: 27 The loss for epoch 3 0.2890692774542037 1 The running loss is: 7.099390174262226 The number of items in train is: 27 The loss for epoch 4 0.26294037682452687 The running loss is: 6.51896614592988 The number of items in train is: 27 The loss for epoch 5 0.24144319058999558
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-summary.json
The running loss is: 6.50308578943077 The number of items in train is: 27 The loss for epoch 6 0.24085502923817667 1 The running loss is: 6.709017640678212 The number of items in train is: 27 The loss for epoch 7 0.24848213483993378 The running loss is: 6.639082251727814 The number of items in train is: 27 The loss for epoch 8 0.2458919352491783 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-summary.json
The running loss is: 5.743373372781207 The number of items in train is: 27 The loss for epoch 9 0.21271753232522989 Data saved to: 28_May_202005_49PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_49PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['xoduj0zf'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 758.606506
66 2020-04-22 sub_region ... 66 838.690002
67 2020-04-23 sub_region ... 67 739.158081
68 2020-04-24 sub_region ... 68 760.182251
69 2020-04-25 sub_region ... 69 791.689392
70 2020-04-26 sub_region ... 70 821.668945
71 2020-04-27 sub_region ... 71 813.191467
72 2020-04-28 sub_region ... 72 824.506714
73 2020-04-29 sub_region ... 73 810.240173
74 2020-04-30 sub_region ... 74 760.039307
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/media/plotly/test_plot_20_50b5ba08.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: xoduj0zf
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-summary.json INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174916-xoduj0zf/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: xoduj0zf INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: br2d7992 with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 1 use_mask: False wandb: Agent Started Run: br2d7992
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/br2d7992 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmJyMmQ3OTkyOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
11.243638909421861
The number of items in train is:
27
The loss for epoch 0
0.41643107071932817
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/media/graph/graph_0_summary_cf4d9cf1.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/media
The running loss is: 15.065432488219813 The number of items in train is: 27 The loss for epoch 1 0.5579789810451783 1 The running loss is: 9.927668149583042 The number of items in train is: 27 The loss for epoch 2 0.36769141294752006
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-summary.json
The running loss is: 7.617937269620597 The number of items in train is: 27 The loss for epoch 3 0.2821458248007629 1 The running loss is: 6.723564739339054 The number of items in train is: 27 The loss for epoch 4 0.2490209162718168 The running loss is: 7.1353028768789954 The number of items in train is: 27 The loss for epoch 5 0.2642704769214443 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-summary.json
The running loss is: 5.726659771054983 The number of items in train is: 27 The loss for epoch 6 0.21209851003907346 2 The running loss is: 6.842962785856798 The number of items in train is: 27 The loss for epoch 7 0.2534430661428444 The running loss is: 6.147264284198172 The number of items in train is: 27 The loss for epoch 8 0.22767645497030267 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-summary.json
The running loss is: 6.772955868189456 The number of items in train is: 27 The loss for epoch 9 0.25085021734035023 Data saved to: 28_May_202005_49PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_49PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['br2d7992'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 702.101074
66 2020-04-22 sub_region ... 66 799.573730
67 2020-04-23 sub_region ... 67 688.857788
68 2020-04-24 sub_region ... 68 695.529663
69 2020-04-25 sub_region ... 69 724.923462
70 2020-04-26 sub_region ... 70 761.772095
71 2020-04-27 sub_region ... 71 748.695679
72 2020-04-28 sub_region ... 72 803.100220
73 2020-04-29 sub_region ... 73 733.273804
74 2020-04-30 sub_region ... 74 718.350464
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/media/plotly/test_plot_20_e049cd21.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: br2d7992
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174929-br2d7992/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: br2d7992 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ftqp9o2w with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 2 use_mask: True wandb: Agent Started Run: ftqp9o2w
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ftqp9o2w INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmZ0cXA5bzJ3OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
13.754779417067766
The number of items in train is:
27
The loss for epoch 0
0.5094362747062136
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/media/graph/graph_0_summary_4d05899a.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/media/graph
The running loss is: 15.335549898445606 The number of items in train is: 27 The loss for epoch 1 0.5679833295720594 1 The running loss is: 12.792411483824253 The number of items in train is: 27 The loss for epoch 2 0.4737930179194168
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-summary.json
The running loss is: 10.463905839249492 The number of items in train is: 27 The loss for epoch 3 0.38755206812035153 1 The running loss is: 8.178016487509012 The number of items in train is: 27 The loss for epoch 4 0.3028894995373708 The running loss is: 7.196833549998701 The number of items in train is: 27 The loss for epoch 5 0.2665493907406926 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-summary.json
The running loss is: 7.724783644080162 The number of items in train is: 27 The loss for epoch 6 0.2861030979288949 2 The running loss is: 7.328360717743635 The number of items in train is: 27 The loss for epoch 7 0.27142076732383835 The running loss is: 8.112628399394453 The number of items in train is: 27 The loss for epoch 8 0.30046771849609083
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-summary.json
1 The running loss is: 6.612887730821967 The number of items in train is: 27 The loss for epoch 9 0.244921767808221 Data saved to: 28_May_202005_49PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['ftqp9o2w']
Data saved to: 28_May_202005_49PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 785.347290
66 2020-04-22 sub_region ... 66 848.155396
67 2020-04-23 sub_region ... 67 778.021973
68 2020-04-24 sub_region ... 68 802.694946
69 2020-04-25 sub_region ... 69 814.782837
70 2020-04-26 sub_region ... 70 827.907227
71 2020-04-27 sub_region ... 71 817.994751
72 2020-04-28 sub_region ... 72 834.097107
73 2020-04-29 sub_region ... 73 812.438049
74 2020-04-30 sub_region ... 74 797.486328
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/media/plotly/test_plot_20_e9ff783e.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ftqp9o2w
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_174945-ftqp9o2w/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ftqp9o2w INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: rugrz0ma with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 2 use_mask: False wandb: Agent Started Run: rugrz0ma
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/rugrz0ma INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnJ1Z3J6MG1hOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
13.334896691143513
The number of items in train is:
27
The loss for epoch 0
0.4938850626349449
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/media/graph/graph_0_summary_54797028.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/media
The running loss is: 15.006994970142841 The number of items in train is: 27 The loss for epoch 1 0.5558146285238089 1 The running loss is: 11.964153863489628 The number of items in train is: 27 The loss for epoch 2 0.44311680975887513
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-summary.json
The running loss is: 9.46366179548204 The number of items in train is: 27 The loss for epoch 3 0.3505059924252607 1 The running loss is: 7.935978587716818 The number of items in train is: 27 The loss for epoch 4 0.29392513287840066 The running loss is: 7.160205790773034 The number of items in train is: 27 The loss for epoch 5 0.26519280706566795 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-summary.json
The running loss is: 7.432532018050551 The number of items in train is: 27 The loss for epoch 6 0.2752789636315019 2 The running loss is: 6.930170862004161 The number of items in train is: 27 The loss for epoch 7 0.256672994889043 The running loss is: 7.583688434679061 The number of items in train is: 27 The loss for epoch 8 0.2808773494325578 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-summary.json
The running loss is: 6.852206656709313 The number of items in train is: 27 The loss for epoch 9 0.2537854317299746 Data saved to: 28_May_202005_50PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_50PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['rugrz0ma']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-history.jsonl
date level ... original_index preds 54 2020-04-10 sub_region ... 54 0.000000 55 2020-04-11 sub_region ... 55 0.000000 56 2020-04-12 sub_region ... 56 0.000000 57 2020-04-13 sub_region ... 57 0.000000 58 2020-04-14 sub_region ... 58 0.000000 59 2020-04-15 sub_region ... 59 0.000000 60 2020-04-16 sub_region ... 60 0.000000 61 2020-04-17 sub_region ... 61 0.000000 62 2020-04-18 sub_region ... 62 0.000000 63 2020-04-19 sub_region ... 63 0.000000 64 2020-04-20 sub_region ... 64 0.000000 65 2020-04-21 sub_region ... 65 792.838745 66 2020-04-22 sub_region ... 66 839.469238 67 2020-04-23 sub_region ... 67 793.361755 68 2020-04-24 sub_region ... 68 806.429504 69 2020-04-25 sub_region ... 69 816.609619 70 2020-04-26 sub_region ... 70 828.707764 71 2020-04-27 sub_region ... 71 823.823608 72 2020-04-28 sub_region ... 72 829.279419 73 2020-04-29 sub_region ... 73 820.153503 74 2020-04-30 sub_region ... 74 806.989746 [21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/media/plotly/test_plot_20_4bc95361.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: rugrz0ma
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175001-rugrz0ma/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: rugrz0ma INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: jheap938 with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 3 use_mask: True wandb: Agent Started Run: jheap938
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/jheap938 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-metadata.json INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmpoZWFwOTM4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-events.jsonl
The running loss is: 18.989742126315832 The number of items in train is: 26 The loss for epoch 0 0.7303746971659936 The running loss is: 16.38419210910797 The number of items in train is: 26 The loss for epoch 1 0.6301612349656912 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/media/graph/graph_0_summary_6bbc3677.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/media/graph
The running loss is: 15.279926925897598 The number of items in train is: 26 The loss for epoch 2 0.5876894971499076 The running loss is: 10.945733822882175 The number of items in train is: 26 The loss for epoch 3 0.4209897624185452 1
INFO:wandb.wandb_agent:Running runs: ['jheap938']
The running loss is: 5.606289468705654 The number of items in train is: 26 The loss for epoch 4 0.21562651802714056 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-history.jsonl
The running loss is: 4.256948810070753 The number of items in train is: 26 The loss for epoch 5 0.16372880038733667 The running loss is: 4.631639575585723 The number of items in train is: 26 The loss for epoch 6 0.17813998367637396 1 The running loss is: 6.9152334509417415 The number of items in train is: 26 The loss for epoch 7 0.26597051734391314
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-history.jsonl
The running loss is: 3.967709585092962 The number of items in train is: 26 The loss for epoch 8 0.15260421481126776 The running loss is: 3.5715523255057633 The number of items in train is: 26 The loss for epoch 9 0.13736739713483706 Data saved to: 28_May_202005_50PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-history.jsonl
Data saved to: 28_May_202005_50PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 814.696899
66 2020-04-22 sub_region ... 66 825.249207
67 2020-04-23 sub_region ... 67 843.944824
68 2020-04-24 sub_region ... 68 816.804077
69 2020-04-25 sub_region ... 69 796.358887
70 2020-04-26 sub_region ... 70 799.779358
71 2020-04-27 sub_region ... 71 801.697021
72 2020-04-28 sub_region ... 72 800.930298
73 2020-04-29 sub_region ... 73 815.445740
74 2020-04-30 sub_region ... 74 822.305115
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: jheap938
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/media/plotly/test_plot_20_d82f34c9.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175018-jheap938/wandb-events.jsonl INFO:wandb.wandb_agent:Cleaning up finished run: jheap938 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 3tx202d6 with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 3 use_mask: False wandb: Agent Started Run: 3tx202d6
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/3tx202d6 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjN0eDIwMmQ2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
18.905425552278757
The number of items in train is:
26
The loss for epoch 0
0.7271317520107214
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/media/graph/graph_0_summary_759919e1.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/media
The running loss is: 16.324342384934425 The number of items in train is: 26 The loss for epoch 1 0.6278593224974779 1 The running loss is: 15.281524695456028 The number of items in train is: 26 The loss for epoch 2 0.5877509498252318 The running loss is: 10.686566263437271 The number of items in train is: 26 The loss for epoch 3 0.41102177936297196
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-summary.json
1 The running loss is: 6.059355471283197 The number of items in train is: 26 The loss for epoch 4 0.2330521335108922 The running loss is: 4.697019984945655 The number of items in train is: 26 The loss for epoch 5 0.1806546148056021
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-summary.json
The running loss is: 6.977002476342022 The number of items in train is: 26 The loss for epoch 6 0.26834624909007776 1 The running loss is: 5.771942357998341 The number of items in train is: 26 The loss for epoch 7 0.2219977829999362 The running loss is: 4.5645317947492 The number of items in train is: 26 The loss for epoch 8 0.17555891518266156
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-summary.json
The running loss is: 5.686972144991159 The number of items in train is: 26 The loss for epoch 9 0.21872969788427538 1 Data saved to: 28_May_202005_50PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_50PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['3tx202d6'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 736.851746
66 2020-04-22 sub_region ... 66 772.893066
67 2020-04-23 sub_region ... 67 762.027588
68 2020-04-24 sub_region ... 68 752.332520
69 2020-04-25 sub_region ... 69 736.720947
70 2020-04-26 sub_region ... 70 741.464417
71 2020-04-27 sub_region ... 71 733.462769
72 2020-04-28 sub_region ... 72 760.639526
73 2020-04-29 sub_region ... 73 746.799805
74 2020-04-30 sub_region ... 74 770.807861
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/media/plotly/test_plot_20_556d85cf.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 3tx202d6
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175034-3tx202d6/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 3tx202d6 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: qoi6pait with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 4 use_mask: True wandb: Agent Started Run: qoi6pait
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/qoi6pait INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnFvaTZwYWl0OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
16.948153391480446
The number of items in train is:
26
The loss for epoch 0
0.6518520535184786
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/media/graph/graph_0_summary_0475200c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/media/graph
The running loss is: 13.85379159078002 The number of items in train is: 26 The loss for epoch 1 0.5328381381069238 The running loss is: 9.64835274219513 The number of items in train is: 26 The loss for epoch 2 0.37109049008442807 1 The running loss is: 8.4732926171273 The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-summary.json
26 The loss for epoch 3 0.3258958698895115 The running loss is: 5.7244476936757565 The number of items in train is: 26 The loss for epoch 4 0.22017106514137524 The running loss is: 6.5513798370957375 The number of items in train is: 26 The loss for epoch 5 0.25197614758060527 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-summary.json
The running loss is: 8.37942417897284 The number of items in train is: 26 The loss for epoch 6 0.3222855453451092 The running loss is: 4.463485406711698 The number of items in train is: 26 The loss for epoch 7 0.1716725156427576 1 The running loss is: 4.722438246943057 The number of items in train is: 26 The loss for epoch 8 0.18163224026704064
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-summary.json
The running loss is: 4.340588949620724 The number of items in train is: 26 The loss for epoch 9 0.1669457288315663 1 Data saved to: 28_May_202005_50PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_50PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['qoi6pait']
interpolate should be below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/config.yaml
Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-summary.json
CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 851.006592
66 2020-04-22 sub_region ... 66 861.678711
67 2020-04-23 sub_region ... 67 870.574524
68 2020-04-24 sub_region ... 68 854.424133
69 2020-04-25 sub_region ... 69 835.877502
70 2020-04-26 sub_region ... 70 840.583374
71 2020-04-27 sub_region ... 71 839.167969
72 2020-04-28 sub_region ... 72 850.335999
73 2020-04-29 sub_region ... 73 848.938354
74 2020-04-30 sub_region ... 74 859.935547
[21 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/media/plotly/test_plot_20_0f4d1d99.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: qoi6pait
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175051-qoi6pait/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: qoi6pait INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: f2z9xtis with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 4 use_mask: False wandb: Agent Started Run: f2z9xtis
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/f2z9xtis INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmYyejl4dGlzOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
16.634709760546684
The number of items in train is:
26
The loss for epoch 0
0.6397965292517955
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/media/graph/graph_0_summary_52394716.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/media/graph
The running loss is: 14.594469293951988 The number of items in train is: 26 The loss for epoch 1 0.5613257420750765 The running loss is: 9.017393708229065 The number of items in train is: 26 The loss for epoch 2 0.3468228349318871 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/wandb-history.jsonl
The running loss is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/wandb-summary.json
7.2338755913078785 The number of items in train is: 26 The loss for epoch 3 0.27822598428107226 The running loss is: 4.978128891438246 The number of items in train is: 26 The loss for epoch 4 0.19146649582454792 1 The running loss is: 5.498803533613682 The number of items in train is: 26 The loss for epoch 5 0.21149244360052621 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/wandb-summary.json
The running loss is: 7.14033218100667 The number of items in train is: 26 The loss for epoch 6 0.27462816080794883 3 Stopping model now Data saved to: 28_May_202005_51PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_51PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 736.526611
66 2020-04-22 sub_region ... 66 758.064575
67 2020-04-23 sub_region ... 67 751.773193
68 2020-04-24 sub_region ... 68 746.766174
69 2020-04-25 sub_region ... 69 738.920166
70 2020-04-26 sub_region ... 70 743.997314
71 2020-04-27 sub_region ... 71 741.321045
72 2020-04-28 sub_region ... 72 750.091064
73 2020-04-29 sub_region ... 73 743.901367
74 2020-04-30 sub_region ... 74 754.618958
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
INFO:wandb.wandb_agent:Running runs: ['f2z9xtis']
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: f2z9xtis
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/media/plotly/test_plot_14_251cab71.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/media/plotly/test_plot_all_15_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175108-f2z9xtis/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: f2z9xtis INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 56hcdtcy with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 5 use_mask: True wandb: Agent Started Run: 56hcdtcy
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/56hcdtcy INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjU2aGNkdGN5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
14.528097435832024
The number of items in train is:
25
The loss for epoch 0
0.581123897433281
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/media/graph/graph_0_summary_a0825f49.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/media/graph
The running loss is: 14.321749061346054 The number of items in train is: 25 The loss for epoch 1 0.5728699624538421 The running loss is: 8.53690830618143 The number of items in train is: 25 The loss for epoch 2 0.3414763322472572 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-history.jsonl
The running loss is: 3.877267986536026 The number of items in train is: 25 The loss for epoch 3 0.15509071946144104 The running loss is: 3.9804319692775607 The number of items in train is: 25 The loss for epoch 4 0.15921727877110242 1 The running loss is: 4.088011069223285 The number of items in train is: 25 The loss for epoch 5 0.16352044276893138
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-history.jsonl
The running loss is: 3.642200954724103 The number of items in train is: 25 The loss for epoch 6 0.14568803818896414 1 The running loss is: 4.692892727442086 The number of items in train is: 25 The loss for epoch 7 0.18771570909768343 2 The running loss is: 3.2474756240844727 The number of items in train is: 25 The loss for epoch 8 0.1298990249633789
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-history.jsonl
The running loss is: 3.19531544810161 The number of items in train is: 25 The loss for epoch 9 0.1278126179240644 Data saved to: 28_May_202005_51PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_51PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['56hcdtcy'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 711.413086
66 2020-04-22 sub_region ... 66 714.489502
67 2020-04-23 sub_region ... 67 749.420166
68 2020-04-24 sub_region ... 68 730.427063
69 2020-04-25 sub_region ... 69 707.012146
70 2020-04-26 sub_region ... 70 700.846252
71 2020-04-27 sub_region ... 71 701.634644
72 2020-04-28 sub_region ... 72 702.439697
73 2020-04-29 sub_region ... 73 710.740356
74 2020-04-30 sub_region ... 74 736.382812
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/media/plotly/test_plot_20_b4b65440.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 56hcdtcy
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175120-56hcdtcy/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 56hcdtcy INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: aiezfxcc with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 1 out_seq_length: 5 use_mask: False wandb: Agent Started Run: aiezfxcc
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/aiezfxcc INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmFpZXpmeGNjOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
14.583468616008759
The number of items in train is:
25
The loss for epoch 0
0.5833387446403503
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/media/graph/graph_0_summary_aa20f3ff.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/media
The running loss is: 11.234198205173016 The number of items in train is: 25 The loss for epoch 1 0.4493679282069206 1 The running loss is: 6.8342317044734955 The number of items in train is: 25 The loss for epoch 2 0.2733692681789398 2 The running loss is: 4.914771635085344 The number of items in train is: 25 The loss for epoch 3 0.19659086540341378
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-summary.json
The running loss is: 6.097113943658769 The number of items in train is: 25 The loss for epoch 4 0.24388455774635076 1 The running loss is: 5.855831284075975 The number of items in train is: 25 The loss for epoch 5 0.23423325136303902 2 The running loss is: 4.823610462248325 The number of items in train is: 25 The loss for epoch 6 0.19294441848993302
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-summary.json
The running loss is: 5.116255223751068 The number of items in train is: 25 The loss for epoch 7 0.2046502089500427 1 The running loss is: 3.9354080138728023 The number of items in train is: 25 The loss for epoch 8 0.1574163205549121 2 The running loss is: 3.635712749324739 The number of items in train is: 25 The loss for epoch 9 0.14542850997298956
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-summary.json
3 Stopping model now Data saved to: 28_May_202005_51PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_51PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['aiezfxcc']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 742.793396
66 2020-04-22 sub_region ... 66 736.652344
67 2020-04-23 sub_region ... 67 754.830688
68 2020-04-24 sub_region ... 68 749.173950
69 2020-04-25 sub_region ... 69 738.829346
70 2020-04-26 sub_region ... 70 734.617004
71 2020-04-27 sub_region ... 71 732.794556
72 2020-04-28 sub_region ... 72 736.974976
73 2020-04-29 sub_region ... 73 737.531616
74 2020-04-30 sub_region ... 74 753.737549
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/media/plotly/test_plot_20_71421788.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: aiezfxcc
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175138-aiezfxcc/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: aiezfxcc INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 01v7tee2 with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 1 use_mask: True wandb: Agent Started Run: 01v7tee2
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/01v7tee2 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjAxdjd0ZWUyOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/media/graph/graph_0_summary_ee2261c4.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/media/graph
The running loss is: 12.349453534625354 The number of items in train is: 27 The loss for epoch 0 0.4573871679490872 The running loss is: 30.535183822968975 The number of items in train is: 27 The loss for epoch 1 1.130932734184036
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-summary.json
The running loss is: 18.304419979453087 The number of items in train is: 27 The loss for epoch 2 0.6779414807204847 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-summary.json
The running loss is: 10.870923589682207 The number of items in train is: 27 The loss for epoch 3 0.4026267996178595 2 The running loss is: 9.898105231113732 The number of items in train is: 27 The loss for epoch 4 0.36659649004124933 3 Stopping model now Data saved to: 28_May_202005_51PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-summary.json
Data saved to: 28_May_202005_51PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
INFO:wandb.wandb_agent:Running runs: ['01v7tee2'] /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 502.529541
66 2020-04-22 sub_region ... 66 492.594849
67 2020-04-23 sub_region ... 67 496.524628
68 2020-04-24 sub_region ... 68 495.569824
69 2020-04-25 sub_region ... 69 495.434570
70 2020-04-26 sub_region ... 70 493.173859
71 2020-04-27 sub_region ... 71 493.890808
72 2020-04-28 sub_region ... 72 491.183136
73 2020-04-29 sub_region ... 73 497.434570
74 2020-04-30 sub_region ... 74 489.901245
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/media/plotly/test_plot_all_11_6655637c.plotly.json
wandb: Agent Finished Run: 01v7tee2
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/media/plotly/test_plot_10_c557f1bd.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/media/plotly INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175154-01v7tee2/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 01v7tee2 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: dziok6s7 with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 1 use_mask: False wandb: Agent Started Run: dziok6s7
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/dziok6s7 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmR6aW9rNnM3OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/media/graph/graph_0_summary_0ec97a73.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/media/graph
The running loss is: 12.00125639885664 The number of items in train is: 27 The loss for epoch 0 0.4444909777354311 The running loss is: 27.564760353183374 The number of items in train is: 27 The loss for epoch 1 1.0209170501179028
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-summary.json
The running loss is: 19.1493139838567 The number of items in train is: 27 The loss for epoch 2 0.7092338512539519 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-summary.json
The running loss is: 9.482985021779314 The number of items in train is: 27 The loss for epoch 3 0.3512216674733079 The running loss is: 8.379846808500588 The number of items in train is: 27 The loss for epoch 4 0.31036469661113286 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-summary.json
The running loss is: 6.564052293615532 The number of items in train is: 27 The loss for epoch 5 0.24311304791168636
INFO:wandb.wandb_agent:Running runs: ['dziok6s7']
The running loss is: 7.043116731874761 The number of items in train is: 27 The loss for epoch 6 0.2608561752546208
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-summary.json
The running loss is: 6.5104550529504195 The number of items in train is: 27 The loss for epoch 7 0.2411279649240896 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-summary.json
The running loss is: 7.045392430853099 The number of items in train is: 27 The loss for epoch 8 0.2609404604019666 The running loss is: 6.926817309111357 The number of items in train is: 27 The loss for epoch 9 0.25654878922634655 1 Data saved to: 28_May_202005_52PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-summary.json
Data saved to: 28_May_202005_52PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 778.043945
66 2020-04-22 sub_region ... 66 783.455933
67 2020-04-23 sub_region ... 67 766.487671
68 2020-04-24 sub_region ... 68 773.625916
69 2020-04-25 sub_region ... 69 782.676147
70 2020-04-26 sub_region ... 70 784.909546
71 2020-04-27 sub_region ... 71 784.748901
72 2020-04-28 sub_region ... 72 778.614258
73 2020-04-29 sub_region ... 73 782.117676
74 2020-04-30 sub_region ... 74 761.960938
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: dziok6s7
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/media/plotly/test_plot_20_647b6c30.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175205-dziok6s7/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: dziok6s7 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: fgsk32jf with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 2 use_mask: True wandb: Agent Started Run: fgsk32jf
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/fgsk32jf INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmZnc2szMmpmOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/media/graph/graph_0_summary_5f629476.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/media/graph
The running loss is: 15.362042212858796 The number of items in train is: 27 The loss for epoch 0 0.5689645264021777 The running loss is: 18.604257106781006 The number of items in train is: 27 The loss for epoch 1 0.6890465595104076 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-history.jsonl
The running loss is: 19.387704892084002 The number of items in train is: 27 The loss for epoch 2 0.7180631441512594
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-history.jsonl
The running loss is: 17.115502156317234 The number of items in train is: 27 The loss for epoch 3 0.6339074872710087 The running loss is: 9.399747041985393 The number of items in train is: 27 The loss for epoch 4 0.3481387793327923 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-history.jsonl
The running loss is: 8.062234118580818 The number of items in train is: 27 The loss for epoch 5 0.2986012636511414
INFO:wandb.wandb_agent:Running runs: ['fgsk32jf']
The running loss is: 7.508644045330584 The number of items in train is: 27 The loss for epoch 6 0.2780979276048364
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-history.jsonl
The running loss is: 7.624856340698898 The number of items in train is: 27 The loss for epoch 7 0.2824020866925518 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-history.jsonl
The running loss is: 7.4580579325556755 The number of items in train is: 27 The loss for epoch 8 0.27622436787243243 The running loss is: 6.958448266610503 The number of items in train is: 27 The loss for epoch 9 0.25772030617075936 Data saved to: 28_May_202005_52PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-history.jsonl
Data saved to: 28_May_202005_52PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 772.555298
66 2020-04-22 sub_region ... 66 766.406433
67 2020-04-23 sub_region ... 67 773.479492
68 2020-04-24 sub_region ... 68 770.044922
69 2020-04-25 sub_region ... 69 772.980347
70 2020-04-26 sub_region ... 70 772.595215
71 2020-04-27 sub_region ... 71 772.935425
72 2020-04-28 sub_region ... 72 770.757202
73 2020-04-29 sub_region ... 73 772.154602
74 2020-04-30 sub_region ... 74 771.334900
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/media/plotly/test_plot_20_cf3ed598.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/media/plotly
wandb: Agent Finished Run: fgsk32jf
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175222-fgsk32jf/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: fgsk32jf INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 1z8e4n5m with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 2 use_mask: False wandb: Agent Started Run: 1z8e4n5m
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/1z8e4n5m INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjF6OGU0bjVtOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/media/graph/graph_0_summary_017f2ad7.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/media/graph
The running loss is: 15.227613762952387 The number of items in train is: 27 The loss for epoch 0 0.5639856949241625 The running loss is: 17.898133425042033 The number of items in train is: 27 The loss for epoch 1 0.6628938305571124 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-history.jsonl
The running loss is: 15.654380347579718 The number of items in train is: 27 The loss for epoch 2 0.5797918647251747
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-history.jsonl
The running loss is: 9.89348728954792 The number of items in train is: 27 The loss for epoch 3 0.3664254551684415 1 The running loss is: 8.431944162817672 The number of items in train is: 27 The loss for epoch 4 0.31229422825250636
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-history.jsonl
The running loss is: 8.609166414942592 The number of items in train is: 27 The loss for epoch 5 0.31885801536824415 1
INFO:wandb.wandb_agent:Running runs: ['1z8e4n5m']
The running loss is: 9.016866168007255 The number of items in train is: 27 The loss for epoch 6 0.3339580062224909 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-history.jsonl
The running loss is: 8.0558643322438 The number of items in train is: 27 The loss for epoch 7 0.29836534563865924
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-history.jsonl
The running loss is: 8.685117572546005 The number of items in train is: 27 The loss for epoch 8 0.3216710212054076 1 The running loss is: 6.745234100148082 The number of items in train is: 27 The loss for epoch 9 0.2498234851906697 Data saved to: 28_May_202005_52PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-history.jsonl
Data saved to: 28_May_202005_52PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 783.742432
66 2020-04-22 sub_region ... 66 782.695801
67 2020-04-23 sub_region ... 67 783.955322
68 2020-04-24 sub_region ... 68 783.325439
69 2020-04-25 sub_region ... 69 783.187744
70 2020-04-26 sub_region ... 70 783.111938
71 2020-04-27 sub_region ... 71 783.345886
72 2020-04-28 sub_region ... 72 782.962280
73 2020-04-29 sub_region ... 73 782.751831
74 2020-04-30 sub_region ... 74 783.127808
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 1z8e4n5m
INFO:wandb.run_manager:shutting down system stats and metadata service
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-events.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/media/plotly/test_plot_20_86f33ef2.plotly.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/media/plotly/test_plot_all_21_6655637c.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/media/plotly
INFO:wandb.run_manager:stopping streaming files and file change observer
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175238-1z8e4n5m/wandb-metadata.json
INFO:wandb.wandb_agent:Cleaning up finished run: 1z8e4n5m
wandb: Network error resolved after 0:00:15.324762, resuming normal operation.
INFO:wandb.wandb_agent:Agent received command: run
INFO:wandb.wandb_agent:Agent starting run with config:
batch_size: 2
forecast_history: 11
lr: 0.002
number_encoder_layers: 2
out_seq_length: 3
use_mask: False
DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: zv56nmxw with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 3 use_mask: False wandb: Agent Started Run: zv56nmxw
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/zv56nmxw INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnp2NTZubXh3OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/media/graph/graph_0_summary_aa917372.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/media/graph
The running loss is: 20.714700505137444 The number of items in train is: 26 The loss for epoch 0 0.7967192501975939 The running loss is: 16.891354955732822 The number of items in train is: 26 The loss for epoch 1 0.6496674982974162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-history.jsonl
The running loss is: 16.139072712510824 The number of items in train is: 26 The loss for epoch 2 0.620733565865801 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-history.jsonl
The running loss is: 15.367559241130948 The number of items in train is: 26 The loss for epoch 3 0.5910599708127288 2 The running loss is: 14.148938469588757 The number of items in train is: 26 The loss for epoch 4 0.5441899411380291 3 Stopping model now Data saved to: 28_May_202005_53PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-history.jsonl
Data saved to: 28_May_202005_53PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 440.134735
66 2020-04-22 sub_region ... 66 439.800903
67 2020-04-23 sub_region ... 67 443.283325
68 2020-04-24 sub_region ... 68 442.842438
69 2020-04-25 sub_region ... 69 440.653381
70 2020-04-26 sub_region ... 70 441.242767
71 2020-04-27 sub_region ... 71 440.577118
72 2020-04-28 sub_region ... 72 442.682800
73 2020-04-29 sub_region ... 73 441.382904
74 2020-04-30 sub_region ... 74 445.453522
[21 rows x 28 columns]
INFO:wandb.wandb_agent:Running runs: ['zv56nmxw']
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/media/plotly/test_plot_10_0d49373a.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: zv56nmxw
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/media/plotly/test_plot_all_11_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175309-zv56nmxw/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: zv56nmxw INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: htbwvcg1 with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 4 use_mask: True wandb: Agent Started Run: htbwvcg1
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/htbwvcg1 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmh0Ynd2Y2cxOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/media/graph/graph_0_summary_41ed4eef.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/media/graph
The running loss is: 23.482627481222153 The number of items in train is: 26 The loss for epoch 0 0.9031779800470059 The running loss is: 20.23513476550579 The number of items in train is: 26 The loss for epoch 1 0.778274414057915 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-summary.json
The running loss is: 21.649361565709114 The number of items in train is: 26 The loss for epoch 2 0.8326677525272737 The running loss is: 15.995433270931244 The number of items in train is: 26 The loss for epoch 3
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-history.jsonl
0.615208971958894
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-summary.json
1 The running loss is: 12.781472019851208 The number of items in train is: 26 The loss for epoch 4 0.4915950776865849 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-summary.json
The running loss is: 8.850842893123627 The number of items in train is: 26 The loss for epoch 5 0.3404170343509087 The running loss is: 6.8902468085289 The number of items in train is: 26 The loss for epoch 6 0.26500949263572693
INFO:wandb.wandb_agent:Running runs: ['htbwvcg1']
1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-summary.json
The running loss is: 8.579151637852192 The number of items in train is: 26 The loss for epoch 7 0.32996737068662274 2 The running loss is: 6.014725340530276 The number of items in train is: 26 The loss for epoch 8 0.23133559002039525
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-summary.json
The running loss is: 4.920796798542142 The number of items in train is: 26 The loss for epoch 9 0.18926141532854393 1 Data saved to: 28_May_202005_53PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-summary.json
Data saved to: 28_May_202005_53PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/config.yaml /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 784.844604
66 2020-04-22 sub_region ... 66 785.061829
67 2020-04-23 sub_region ... 67 785.951660
68 2020-04-24 sub_region ... 68 785.422729
69 2020-04-25 sub_region ... 69 784.747925
70 2020-04-26 sub_region ... 70 784.533691
71 2020-04-27 sub_region ... 71 784.408447
72 2020-04-28 sub_region ... 72 785.171936
73 2020-04-29 sub_region ... 73 784.649170
74 2020-04-30 sub_region ... 74 785.943665
[21 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/media/plotly/test_plot_20_bf33d510.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: htbwvcg1
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175321-htbwvcg1/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: htbwvcg1 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: frnme9lu with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 4 use_mask: False wandb: Agent Started Run: frnme9lu
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/frnme9lu INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmZybm1lOWx1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/media/graph/graph_0_summary_649639ab.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/media/graph
The running loss is: 23.31422958523035 The number of items in train is: 26 The loss for epoch 0 0.896701137893475 The running loss is: 19.628748193383217 The number of items in train is: 26 The loss for epoch 1 0.7549518535916622 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-history.jsonl
The running loss is: 22.03840383887291 The number of items in train is: 26 The loss for epoch 2 0.8476309168797272 The running loss is: 15.539160393178463 The number of items in train is: 26 The loss for epoch 3 0.5976600151222485
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-history.jsonl
1 The running loss is: 12.818177118897438 The number of items in train is: 26 The loss for epoch 4 0.4930068122652861
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-history.jsonl
The running loss is: 10.63734956085682 The number of items in train is: 26 The loss for epoch 5 0.4091288292637238 1 The running loss is: 6.949803344905376 The number of items in train is: 26 The loss for epoch 6 0.2673001286502068
INFO:wandb.wandb_agent:Running runs: ['frnme9lu'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-history.jsonl
The running loss is: 7.64347056671977 The number of items in train is: 26 The loss for epoch 7 0.29397963718152964 1 The running loss is: 11.507175385951996 The number of items in train is: 26 The loss for epoch 8 0.4425836686904614
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-history.jsonl
The running loss is: 5.667488571256399 The number of items in train is: 26 The loss for epoch 9 0.21798032966370767 Data saved to: 28_May_202005_53PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_53PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 749.537842
66 2020-04-22 sub_region ... 66 753.229736
67 2020-04-23 sub_region ... 67 753.515015
68 2020-04-24 sub_region ... 68 753.205444
69 2020-04-25 sub_region ... 69 751.122925
70 2020-04-26 sub_region ... 70 750.473755
71 2020-04-27 sub_region ... 71 748.162842
72 2020-04-28 sub_region ... 72 752.245361
73 2020-04-29 sub_region ... 73 750.474243
74 2020-04-30 sub_region ... 74 754.320679
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/media/plotly/test_plot_20_67e12e5e.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: frnme9lu
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175338-frnme9lu/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: frnme9lu INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 69e2yc72 with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 5 use_mask: True wandb: Agent Started Run: 69e2yc72
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/69e2yc72 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjY5ZTJ5YzcyOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/media/graph/graph_0_summary_c2fc5416.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/media/graph
The running loss is: 17.830005079507828 The number of items in train is: 25 The loss for epoch 0 0.7132002031803131 The running loss is: 16.9639795422554 The number of items in train is: 25 The loss for epoch 1 0.6785591816902161 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-history.jsonl
The running loss is: 19.417406290769577 The number of items in train is: 25 The loss for epoch 2 0.776696251630783 The running loss is: 13.737709395587444 The number of items in train is: 25 The loss for epoch 3 0.5495083758234978 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-history.jsonl
The running loss is: 13.103170596063137 The number of items in train is: 25 The loss for epoch 4 0.5241268238425255 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-history.jsonl
The running loss is: 9.996175795793533 The number of items in train is: 25 The loss for epoch 5 0.3998470318317413 3 Stopping model now Data saved to: 28_May_202005_53PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_53PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['69e2yc72'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 417.858063
66 2020-04-22 sub_region ... 66 419.026245
67 2020-04-23 sub_region ... 67 420.017090
68 2020-04-24 sub_region ... 68 420.572571
69 2020-04-25 sub_region ... 69 419.899597
70 2020-04-26 sub_region ... 70 419.641113
71 2020-04-27 sub_region ... 71 419.378143
72 2020-04-28 sub_region ... 72 419.695251
73 2020-04-29 sub_region ... 73 419.175232
74 2020-04-30 sub_region ... 74 421.516815
[21 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/media/plotly/test_plot_12_d17e53a2.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 69e2yc72
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/media/plotly/test_plot_all_13_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175354-69e2yc72/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 69e2yc72 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 23bl3fhm with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 2 out_seq_length: 5 use_mask: False wandb: Agent Started Run: 23bl3fhm
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/23bl3fhm INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjIzYmwzZmhtOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/media/graph/graph_0_summary_16c8da2c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/media/graph
The running loss is: 17.741086795926094 The number of items in train is: 25 The loss for epoch 0 0.7096434718370438 The running loss is: 16.730679839849472 The number of items in train is: 25 The loss for epoch 1 0.6692271935939789 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-summary.json
The running loss is: 19.939407482743263 The number of items in train is: 25 The loss for epoch 2 0.7975762993097305 The running loss is: 13.255669929087162 The number of items in train is: 25 The loss for epoch 3 0.5302267971634865
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-summary.json
1 The running loss is: 11.027616888284683 The number of items in train is: 25 The loss for epoch 4 0.4411046755313873 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-summary.json
The running loss is: 6.718288138508797 The number of items in train is: 25 The loss for epoch 5 0.26873152554035185
INFO:wandb.wandb_agent:Running runs: ['23bl3fhm']
The running loss is: 7.313029149547219 The number of items in train is: 25 The loss for epoch 6 0.29252116598188876 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-summary.json
The running loss is: 12.773601979017258 The number of items in train is: 25 The loss for epoch 7 0.5109440791606903 The running loss is: 6.80379575304687 The number of items in train is: 25 The loss for epoch 8 0.2721518301218748 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-summary.json
The running loss is: 4.739170776680112 The number of items in train is: 25 The loss for epoch 9 0.18956683106720448 2 Data saved to: 28_May_202005_54PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_54PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 770.853760
66 2020-04-22 sub_region ... 66 769.976440
67 2020-04-23 sub_region ... 67 771.220154
68 2020-04-24 sub_region ... 68 770.911987
69 2020-04-25 sub_region ... 69 770.517700
70 2020-04-26 sub_region ... 70 770.186768
71 2020-04-27 sub_region ... 71 770.193970
72 2020-04-28 sub_region ... 72 770.157593
73 2020-04-29 sub_region ... 73 770.232788
74 2020-04-30 sub_region ... 74 770.960083
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/media/plotly/test_plot_20_57ec849c.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 23bl3fhm
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175411-23bl3fhm/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 23bl3fhm INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: vslypx7u with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 1 use_mask: True wandb: Agent Started Run: vslypx7u
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/vslypx7u INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnZzbHlweDd1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/media/graph/graph_0_summary_775214d5.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/media
The running loss is: 13.64892402663827 The number of items in train is: 27 The loss for epoch 0 0.5055157046903063
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-summary.json
The running loss is: 37.692751033231616 The number of items in train is: 27 The loss for epoch 1 1.3960278160456154
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-summary.json
The running loss is: 20.47567129507661 The number of items in train is: 27 The loss for epoch 2 0.7583581961139485
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-summary.json
The running loss is: 20.933722829446197 The number of items in train is: 27 The loss for epoch 3 0.7753230677572666 1
INFO:wandb.wandb_agent:Running runs: ['vslypx7u'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-summary.json
The running loss is: 20.798208009451628 The number of items in train is: 27 The loss for epoch 4 0.7703040003500603 The running loss is: 20.184429235756397 The number of items in train is: 27 The loss for epoch 5 0.7475714531761629
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-summary.json
1 The running loss is: 20.411993622779846 The number of items in train is: 27 The loss for epoch 6 0.755999763806661 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-summary.json
The running loss is: 19.94804622977972 The number of items in train is: 27 The loss for epoch 7 0.7388165270288786 3 Stopping model now Data saved to: 28_May_202005_54PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-summary.json
Data saved to: 28_May_202005_54PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 474.087341
66 2020-04-22 sub_region ... 66 474.107361
67 2020-04-23 sub_region ... 67 473.755005
68 2020-04-24 sub_region ... 68 474.003113
69 2020-04-25 sub_region ... 69 474.199402
70 2020-04-26 sub_region ... 70 474.221405
71 2020-04-27 sub_region ... 71 474.191833
72 2020-04-28 sub_region ... 72 473.996887
73 2020-04-29 sub_region ... 73 474.185181
74 2020-04-30 sub_region ... 74 473.697998
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/media/plotly/test_plot_16_1f8d7214.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: vslypx7u
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/media/plotly/test_plot_all_17_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175427-vslypx7u/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: vslypx7u INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 8eb1h9re with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 1 use_mask: False wandb: Agent Started Run: 8eb1h9re
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/8eb1h9re INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjhlYjFoOXJlOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/media/graph/graph_0_summary_eaee8e05.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/media/graph
The running loss is: 13.37478191871196 The number of items in train is: 27 The loss for epoch 0 0.49536229328562814
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-history.jsonl
The running loss is: 37.45775743201375 The number of items in train is: 27 The loss for epoch 1 1.3873243493338425
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-history.jsonl
The running loss is: 20.595209177583456 The number of items in train is: 27 The loss for epoch 2 0.7627855250956835
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-history.jsonl
The running loss is: 21.006036780774593 The number of items in train is: 27 The loss for epoch 3 0.7780013622509109
INFO:wandb.wandb_agent:Running runs: ['8eb1h9re'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-history.jsonl
The running loss is: 20.73020259477198 The number of items in train is: 27 The loss for epoch 4 0.7677852812878512
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-history.jsonl
The running loss is: 20.24301341921091 The number of items in train is: 27 The loss for epoch 5 0.7497412377485523 1 The running loss is: 20.167157787829638 The number of items in train is: 27 The loss for epoch 6 0.7469317699196162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-history.jsonl
2 The running loss is: 18.895125970244408 The number of items in train is: 27 The loss for epoch 7 0.6998194803794225 3 Stopping model now Data saved to: 28_May_202005_54PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-history.jsonl
Data saved to: 28_May_202005_54PM_model.pth interpolate should be below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/config.yaml
Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 474.395355
66 2020-04-22 sub_region ... 66 474.474243
67 2020-04-23 sub_region ... 67 473.946503
68 2020-04-24 sub_region ... 68 474.251740
69 2020-04-25 sub_region ... 69 474.498688
70 2020-04-26 sub_region ... 70 474.531128
71 2020-04-27 sub_region ... 71 474.499969
72 2020-04-28 sub_region ... 72 474.257050
73 2020-04-29 sub_region ... 73 474.527222
74 2020-04-30 sub_region ... 74 473.818420
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/media/plotly/test_plot_16_c257944a.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 8eb1h9re
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/media/plotly/test_plot_all_17_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175444-8eb1h9re/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 8eb1h9re INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: u5disclf with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 2 use_mask: True wandb: Agent Started Run: u5disclf
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/u5disclf INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnU1ZGlzY2xmOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/media/graph/graph_0_summary_6c6e54e0.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/media
The running loss is: 15.137208757922053 The number of items in train is: 27 The loss for epoch 0 0.5606373614045205
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-history.jsonl
The running loss is: 30.88287841156125 The number of items in train is: 27 The loss for epoch 1 1.1438103115393057
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-history.jsonl
The running loss is: 25.371903955936432 The number of items in train is: 27 The loss for epoch 2 0.9397001465161642
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-history.jsonl
The running loss is: 20.891498629003763 The number of items in train is: 27 The loss for epoch 3 0.7737592084816208
INFO:wandb.wandb_agent:Running runs: ['u5disclf'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-history.jsonl
The running loss is: 20.58536821603775 The number of items in train is: 27 The loss for epoch 4 0.7624210450384352 1 The running loss is: 21.16692252457142 The number of items in train is: 27 The loss for epoch 5 0.7839600935026452 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-history.jsonl
The running loss is: 20.775313809514046 The number of items in train is: 27 The loss for epoch 6 0.7694560670190387 3 Stopping model now Data saved to: 28_May_202005_55PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-history.jsonl
Data saved to: 28_May_202005_55PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 422.063049
66 2020-04-22 sub_region ... 66 422.065277
67 2020-04-23 sub_region ... 67 422.008148
68 2020-04-24 sub_region ... 68 422.044128
69 2020-04-25 sub_region ... 69 422.075989
70 2020-04-26 sub_region ... 70 422.085114
71 2020-04-27 sub_region ... 71 422.084900
72 2020-04-28 sub_region ... 72 422.078888
73 2020-04-29 sub_region ... 73 422.060028
74 2020-04-30 sub_region ... 74 422.037537
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: u5disclf
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/media/plotly/test_plot_all_15_6655637c.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/media/plotly/test_plot_14_7a77c25e.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175501-u5disclf/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: u5disclf INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: zpwwfeyr with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 2 use_mask: False wandb: Agent Started Run: zpwwfeyr
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/zpwwfeyr INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnpwd3dmZXlyOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/media/graph/graph_0_summary_97b63218.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/media/graph
The running loss is: 15.071377269923687 The number of items in train is: 27 The loss for epoch 0 0.5581991581453217
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-summary.json
The running loss is: 31.568736284971237 The number of items in train is: 27 The loss for epoch 1 1.1692124549989347
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-summary.json
The running loss is: 25.198651179671288 The number of items in train is: 27 The loss for epoch 2 0.9332833770248625
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-summary.json
The running loss is: 21.102718703448772 The number of items in train is: 27 The loss for epoch 3 0.7815821742018064
INFO:wandb.wandb_agent:Running runs: ['zpwwfeyr'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-summary.json
The running loss is: 20.57193885743618 The number of items in train is: 27 The loss for epoch 4 0.7619236613865252 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-summary.json
The running loss is: 21.114420972764492 The number of items in train is: 27 The loss for epoch 5 0.7820155915838701 2 The running loss is: 21.067522056400776 The number of items in train is: 27 The loss for epoch 6 0.7802785946815102
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-history.jsonl
3
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-summary.json
Stopping model now Data saved to: 28_May_202005_55PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_55PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 423.483063
66 2020-04-22 sub_region ... 66 423.511169
67 2020-04-23 sub_region ... 67 423.472900
68 2020-04-24 sub_region ... 68 423.498932
69 2020-04-25 sub_region ... 69 423.520294
70 2020-04-26 sub_region ... 70 423.524902
71 2020-04-27 sub_region ... 71 423.520996
72 2020-04-28 sub_region ... 72 423.510651
73 2020-04-29 sub_region ... 73 423.502411
74 2020-04-30 sub_region ... 74 423.487152
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/media/plotly/test_plot_14_0bb7c432.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: zpwwfeyr
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/media/plotly/test_plot_all_15_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175519-zpwwfeyr/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: zpwwfeyr INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: b6hcpq3h with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 3 use_mask: True wandb: Agent Started Run: b6hcpq3h
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/b6hcpq3h INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmI2aGNwcTNoOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/media/graph/graph_0_summary_13555ea6.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/media
The running loss is: 24.238217145204544 The number of items in train is: 26 The loss for epoch 0 0.9322391209694055
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-history.jsonl
The running loss is: 16.11776690930128 The number of items in train is: 26 The loss for epoch 1 0.6199141118962032 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-history.jsonl
The running loss is: 16.947837613523006 The number of items in train is: 26 The loss for epoch 2 0.6518399082124233 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-history.jsonl
The running loss is: 17.64614214003086 The number of items in train is: 26 The loss for epoch 3 0.6786977746165715
INFO:wandb.wandb_agent:Running runs: ['b6hcpq3h']
The running loss is: 16.301386021077633 The number of items in train is: 26 The loss for epoch 4 0.6269763854260628
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-history.jsonl
The running loss is: 16.042810605838895 The number of items in train is: 26 The loss for epoch 5 0.6170311771476498
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-history.jsonl
The running loss is: 15.744363751262426 The number of items in train is: 26 The loss for epoch 6 0.6055524519716318 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-history.jsonl
The running loss is: 16.016306111589074 The number of items in train is: 26 The loss for epoch 7 0.6160117735226567 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-history.jsonl
The running loss is: 15.813427444547415 The number of items in train is: 26 The loss for epoch 8 0.6082087478672082
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-history.jsonl
The running loss is: 16.24589281156659 The number of items in train is: 26 The loss for epoch 9 0.6248420312140996 Data saved to: 28_May_202005_55PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_55PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 509.934052
66 2020-04-22 sub_region ... 66 507.839417
67 2020-04-23 sub_region ... 67 509.408234
68 2020-04-24 sub_region ... 68 509.694031
69 2020-04-25 sub_region ... 69 509.575317
70 2020-04-26 sub_region ... 70 508.830750
71 2020-04-27 sub_region ... 71 509.065674
72 2020-04-28 sub_region ... 72 507.174622
73 2020-04-29 sub_region ... 73 509.588074
74 2020-04-30 sub_region ... 74 507.856110
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/media/plotly/test_plot_20_75179c3f.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: b6hcpq3h
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175540-b6hcpq3h/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: b6hcpq3h INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ryhzse7k with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 3 use_mask: False wandb: Agent Started Run: ryhzse7k
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ryhzse7k INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnJ5aHpzZTdrOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/media/graph/graph_0_summary_24bc68c5.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/media/graph
The running loss is: 24.46081606298685 The number of items in train is: 26 The loss for epoch 0 0.9408006178071866
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-summary.json
The running loss is: 16.266936726868153 The number of items in train is: 26 The loss for epoch 1 0.625651412571852 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-summary.json
The running loss is: 16.894336327910423 The number of items in train is: 26 The loss for epoch 2 0.6497821664580932
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-summary.json
The running loss is: 17.256735399365425 The number of items in train is: 26 The loss for epoch 3 0.6637205922832856
INFO:wandb.wandb_agent:Running runs: ['ryhzse7k'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-summary.json
The running loss is: 16.231802832335234 The number of items in train is: 26 The loss for epoch 4 0.6243001089359705
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-history.jsonl
The running loss is: 16.1003286074847
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-summary.json
The number of items in train is: 26 The loss for epoch 5 0.6192434079801807 The running loss is: 15.810061607509851 The number of items in train is: 26 The loss for epoch 6 0.6080792925965327 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-summary.json
The running loss is: 16.048187194392085 The number of items in train is: 26 The loss for epoch 7 0.6172379690150802 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-summary.json
The running loss is: 15.809576485306025 The number of items in train is: 26 The loss for epoch 8 0.6080606340502317 3 Stopping model now Data saved to: 28_May_202005_56PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-summary.json
Data saved to: 28_May_202005_56PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 529.040771
66 2020-04-22 sub_region ... 66 528.143555
67 2020-04-23 sub_region ... 67 528.838806
68 2020-04-24 sub_region ... 68 528.961365
69 2020-04-25 sub_region ... 69 528.801025
70 2020-04-26 sub_region ... 70 528.515015
71 2020-04-27 sub_region ... 71 528.581177
72 2020-04-28 sub_region ... 72 528.077698
73 2020-04-29 sub_region ... 73 528.772461
74 2020-04-30 sub_region ... 74 528.409302
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ryhzse7k
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/media/plotly/test_plot_18_712b0439.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/media/plotly/test_plot_all_19_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175604-ryhzse7k/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ryhzse7k INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 17i2vd1z with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 4 use_mask: True wandb: Agent Started Run: 17i2vd1z
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/17i2vd1z INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjE3aTJ2ZDF6OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/media/graph/graph_0_summary_3528a6ee.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/media
The running loss is: 24.921816915273666 The number of items in train is: 26 The loss for epoch 0 0.958531419818218
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-summary.json
The running loss is: 22.33100463449955 The number of items in train is: 26 The loss for epoch 1 0.858884793634598
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-summary.json
The running loss is: 20.428308714181185 The number of items in train is: 26 The loss for epoch 2 0.7857041813146609
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-summary.json
The running loss is: 17.427796185016632 The number of items in train is: 26 The loss for epoch 3 0.6702998532698705 1
INFO:wandb.wandb_agent:Running runs: ['17i2vd1z']
The running loss is: 17.2829820625484 The number of items in train is: 26 The loss for epoch 4 0.6647300793287846 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-summary.json
The running loss is: 15.887216325849295 The number of items in train is: 26 The loss for epoch 5 0.6110467817634344 3 Stopping model now Data saved to: 28_May_202005_56PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-summary.json
Data saved to: 28_May_202005_56PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 559.779724
66 2020-04-22 sub_region ... 66 560.009094
67 2020-04-23 sub_region ... 67 560.145813
68 2020-04-24 sub_region ... 68 560.165466
69 2020-04-25 sub_region ... 69 560.001709
70 2020-04-26 sub_region ... 70 560.061584
71 2020-04-27 sub_region ... 71 559.992859
72 2020-04-28 sub_region ... 72 560.228088
73 2020-04-29 sub_region ... 73 559.984314
74 2020-04-30 sub_region ... 74 560.468567
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/media/plotly/test_plot_12_f72c3f2f.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 17i2vd1z
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/media/plotly/test_plot_all_13_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175626-17i2vd1z/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 17i2vd1z INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: b81wjoof with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 4 use_mask: False wandb: Agent Started Run: b81wjoof
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/b81wjoof INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmI4MXdqb29mOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/media/graph/graph_0_summary_cba168a9.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/media
The running loss is: 25.268363758921623 The number of items in train is: 26 The loss for epoch 0 0.9718601445739086
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-summary.json
The running loss is: 22.410943418741226 The number of items in train is: 26 The loss for epoch 1 0.8619593622592779
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-summary.json
The running loss is: 20.349964793771505 The number of items in train is: 26 The loss for epoch 2 0.7826909536065964
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-summary.json
The running loss is: 17.21460761502385 The number of items in train is: 26 The loss for epoch 3 0.6621002928855327 1
INFO:wandb.wandb_agent:Running runs: ['b81wjoof']
The running loss is: 17.72370159626007 The number of items in train is: 26 The loss for epoch 4 0.6816808306253873 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-summary.json
The running loss is: 15.540141738951206 The number of items in train is: 26 The loss for epoch 5 0.597697759190431 3 Stopping model now Data saved to: 28_May_202005_56PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-summary.json
Data saved to: 28_May_202005_56PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 559.067444
66 2020-04-22 sub_region ... 66 559.143799
67 2020-04-23 sub_region ... 67 559.442505
68 2020-04-24 sub_region ... 68 559.442566
69 2020-04-25 sub_region ... 69 559.225464
70 2020-04-26 sub_region ... 70 559.264404
71 2020-04-27 sub_region ... 71 559.198975
72 2020-04-28 sub_region ... 72 559.425171
73 2020-04-29 sub_region ... 73 559.358643
74 2020-04-30 sub_region ... 74 559.806274
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: b81wjoof
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/media/plotly/test_plot_12_b8d9f787.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/media/plotly/test_plot_all_13_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175642-b81wjoof/wandb-events.jsonl INFO:wandb.wandb_agent:Cleaning up finished run: b81wjoof INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 2kziz1z8 with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 5 use_mask: True wandb: Agent Started Run: 2kziz1z8
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/2kziz1z8 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjJreml6MXo4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/media/graph/graph_0_summary_d6549107.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/media
The running loss is: 20.4773468375206 The number of items in train is: 25 The loss for epoch 0 0.819093873500824
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-history.jsonl
The running loss is: 19.37137946486473 The number of items in train is: 25 The loss for epoch 1 0.7748551785945892
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-history.jsonl
The running loss is: 16.317528434097767 The number of items in train is: 25 The loss for epoch 2 0.6527011373639107
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-history.jsonl
The running loss is: 14.260967433452606 The number of items in train is: 25 The loss for epoch 3 0.5704386973381043 1
INFO:wandb.wandb_agent:Running runs: ['2kziz1z8']
The running loss is: 14.17140232771635 The number of items in train is: 25 The loss for epoch 4 0.566856093108654
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-history.jsonl
The running loss is: 14.054600022733212 The number of items in train is: 25 The loss for epoch 5 0.5621840009093284
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-history.jsonl
The running loss is: 14.07607826590538 The number of items in train is: 25 The loss for epoch 6 0.5630431306362152
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-history.jsonl
The running loss is: 14.025826334953308 The number of items in train is: 25 The loss for epoch 7 0.5610330533981324 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-history.jsonl
The running loss is: 13.500305883586407 The number of items in train is: 25 The loss for epoch 8 0.5400122353434562 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-history.jsonl
The running loss is: 12.7346723228693 The number of items in train is: 25 The loss for epoch 9 0.509386892914772 3 Stopping model now Data saved to: 28_May_202005_57PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_57PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/config.yaml
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-history.jsonl
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 384.512299
66 2020-04-22 sub_region ... 66 383.916504
67 2020-04-23 sub_region ... 67 384.334045
68 2020-04-24 sub_region ... 68 384.549500
69 2020-04-25 sub_region ... 69 384.431915
70 2020-04-26 sub_region ... 70 384.146332
71 2020-04-27 sub_region ... 71 384.198578
72 2020-04-28 sub_region ... 72 383.738586
73 2020-04-29 sub_region ... 73 384.381348
74 2020-04-30 sub_region ... 74 383.990662
[21 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/media/plotly/test_plot_20_1987c841.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 2kziz1z8
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175659-2kziz1z8/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 2kziz1z8 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: jz1lf7ky with config: batch_size: 2 forecast_history: 11 lr: 0.002 number_encoder_layers: 3 out_seq_length: 5 use_mask: False wandb: Agent Started Run: jz1lf7ky
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/jz1lf7ky INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmp6MWxmN2t5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/media/graph/graph_0_summary_ff747659.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/media/graph
The running loss is: 20.519908547401428 The number of items in train is: 25 The loss for epoch 0 0.8207963418960571
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-summary.json
The running loss is: 19.348501980304718 The number of items in train is: 25 The loss for epoch 1 0.7739400792121888
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-summary.json
The running loss is: 16.020245015621185 The number of items in train is: 25 The loss for epoch 2 0.6408098006248474 The running loss is: 14.293965496122837 The number of items in train is: 25 The loss for epoch 3 0.5717586198449135
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-summary.json
1
INFO:wandb.wandb_agent:Running runs: ['jz1lf7ky']
The running loss is: 14.129456043243408 The number of items in train is: 25 The loss for epoch 4 0.5651782417297363 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-summary.json
The running loss is: 13.794427633285522 The number of items in train is: 25 The loss for epoch 5 0.5517771053314209
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-summary.json
The running loss is: 13.02604092657566 The number of items in train is: 25 The loss for epoch 6 0.5210416370630264 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-summary.json
The running loss is: 9.591085441410542 The number of items in train is: 25 The loss for epoch 7 0.3836434176564217 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-summary.json
The running loss is: 7.496750168502331 The number of items in train is: 25 The loss for epoch 8 0.29987000674009323 3 Stopping model now Data saved to: 28_May_202005_57PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_57PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 388.992981
66 2020-04-22 sub_region ... 66 387.531372
67 2020-04-23 sub_region ... 67 388.493744
68 2020-04-24 sub_region ... 68 388.772583
69 2020-04-25 sub_region ... 69 388.547852
70 2020-04-26 sub_region ... 70 387.934692
71 2020-04-27 sub_region ... 71 388.074402
72 2020-04-28 sub_region ... 72 387.295563
73 2020-04-29 sub_region ... 73 388.444427
74 2020-04-30 sub_region ... 74 387.736328
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/media/plotly/test_plot_18_166901d4.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: jz1lf7ky
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/media/plotly/test_plot_all_19_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175716-jz1lf7ky/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: jz1lf7ky INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: s4fonam9 with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 1 use_mask: True wandb: Agent Started Run: s4fonam9
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/s4fonam9 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnM0Zm9uYW05OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
11.178941723890603
The number of items in train is:
27
The loss for epoch 0
0.4140348786626149
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/media/graph/graph_0_summary_1f3837d4.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/media/graph
The running loss is: 14.482956442050636 The number of items in train is: 27 The loss for epoch 1 0.5364057941500235 1 The running loss is: 12.16446726070717 The number of items in train is: 27 The loss for epoch 2 0.4505358244706359
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-history.jsonl
The running loss is: 7.8048704912635 The number of items in train is: 27 The loss for epoch 3 0.2890692774542037 1 The running loss is: 7.099390174262226 The number of items in train is: 27 The loss for epoch 4 0.26294037682452687 The running loss is: 6.51896614592988 The number of items in train is: 27 The loss for epoch 5 0.24144319058999558
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-history.jsonl
The running loss is: 6.50308578943077 The number of items in train is: 27 The loss for epoch 6 0.24085502923817667 1 The running loss is: 6.709017640678212 The number of items in train is: 27 The loss for epoch 7 0.24848213483993378 The running loss is: 6.639082251727814 The number of items in train is: 27 The loss for epoch 8 0.2458919352491783 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-history.jsonl
The running loss is: 5.743373372781207 The number of items in train is: 27 The loss for epoch 9 0.21271753232522989 Data saved to: 28_May_202005_57PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['s4fonam9']
Data saved to: 28_May_202005_57PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 758.606506
66 2020-04-22 sub_region ... 66 838.690002
67 2020-04-23 sub_region ... 67 739.158081
68 2020-04-24 sub_region ... 68 760.182251
69 2020-04-25 sub_region ... 69 791.689392
70 2020-04-26 sub_region ... 70 821.668945
71 2020-04-27 sub_region ... 71 813.191467
72 2020-04-28 sub_region ... 72 824.506714
73 2020-04-29 sub_region ... 73 810.240173
74 2020-04-30 sub_region ... 74 760.039307
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/media/plotly/test_plot_20_50b5ba08.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: s4fonam9
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175732-s4fonam9/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: s4fonam9 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ay7p9npe with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 1 use_mask: False wandb: Agent Started Run: ay7p9npe
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ay7p9npe INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmF5N3A5bnBlOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/config.yaml INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-metadata.json
interpolate should be below
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-events.jsonl
Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
11.243638909421861
The number of items in train is:
27
The loss for epoch 0
0.41643107071932817
The running loss is:
15.065432488219813
The number of items in train is:
27
The loss for epoch 1
0.5579789810451783
1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/media/graph/graph_0_summary_dec9c9e3.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/media/graph
The running loss is: 9.927668149583042 The number of items in train is: 27 The loss for epoch 2 0.36769141294752006 The running loss is: 7.617937269620597 The number of items in train is: 27 The loss for epoch 3 0.2821458248007629 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-history.jsonl
The running loss is: 6.723564739339054 The number of items in train is: 27 The loss for epoch 4 0.2490209162718168 The running loss is: 7.1353028768789954 The number of items in train is: 27 The loss for epoch 5 0.2642704769214443 1 The running loss is: 5.726659771054983 The number of items in train is: 27 The loss for epoch 6 0.21209851003907346 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-history.jsonl
The running loss is: 6.842962785856798 The number of items in train is: 27 The loss for epoch 7 0.2534430661428444 The running loss is: 6.147264284198172 The number of items in train is: 27 The loss for epoch 8 0.22767645497030267 1
INFO:wandb.wandb_agent:Running runs: ['ay7p9npe']
The running loss is: 6.772955868189456 The number of items in train is: 27 The loss for epoch 9 0.25085021734035023 Data saved to: 28_May_202005_57PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-history.jsonl
Data saved to: 28_May_202005_57PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 702.101074
66 2020-04-22 sub_region ... 66 799.573730
67 2020-04-23 sub_region ... 67 688.857788
68 2020-04-24 sub_region ... 68 695.529663
69 2020-04-25 sub_region ... 69 724.923462
70 2020-04-26 sub_region ... 70 761.772095
71 2020-04-27 sub_region ... 71 748.695679
72 2020-04-28 sub_region ... 72 803.100220
73 2020-04-29 sub_region ... 73 733.273804
74 2020-04-30 sub_region ... 74 718.350464
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/media/plotly/test_plot_20_e049cd21.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ay7p9npe
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175749-ay7p9npe/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ay7p9npe INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: xhybkiq8 with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 2 use_mask: True wandb: Agent Started Run: xhybkiq8
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/xhybkiq8 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnhoeWJraXE4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
13.754779417067766
The number of items in train is:
27
The loss for epoch 0
0.5094362747062136
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/media/graph/graph_0_summary_27ea2e80.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/media
The running loss is: 15.335549898445606 The number of items in train is: 27 The loss for epoch 1 0.5679833295720594 1 The running loss is: 12.792411483824253 The number of items in train is: 27 The loss for epoch 2 0.4737930179194168
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-summary.json
The running loss is: 10.463905839249492 The number of items in train is: 27 The loss for epoch 3 0.38755206812035153 1 The running loss is: 8.178016487509012 The number of items in train is: 27 The loss for epoch 4 0.3028894995373708 The running loss is: 7.196833549998701 The number of items in train is: 27 The loss for epoch 5 0.2665493907406926 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-summary.json
The running loss is: 7.724783644080162 The number of items in train is: 27 The loss for epoch 6 0.2861030979288949 2 The running loss is: 7.328360717743635 The number of items in train is: 27 The loss for epoch 7 0.27142076732383835 The running loss is: 8.112628399394453 The number of items in train is: 27 The loss for epoch 8 0.30046771849609083 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-summary.json
The running loss is: 6.612887730821967 The number of items in train is: 27 The loss for epoch 9 0.244921767808221 Data saved to: 28_May_202005_58PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['xhybkiq8']
Data saved to: 28_May_202005_58PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 785.347290
66 2020-04-22 sub_region ... 66 848.155396
67 2020-04-23 sub_region ... 67 778.021973
68 2020-04-24 sub_region ... 68 802.694946
69 2020-04-25 sub_region ... 69 814.782837
70 2020-04-26 sub_region ... 70 827.907227
71 2020-04-27 sub_region ... 71 817.994751
72 2020-04-28 sub_region ... 72 834.097107
73 2020-04-29 sub_region ... 73 812.438049
74 2020-04-30 sub_region ... 74 797.486328
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/media/plotly/test_plot_20_e9ff783e.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: xhybkiq8
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175806-xhybkiq8/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: xhybkiq8 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 2bp9e1u9 with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 2 use_mask: False wandb: Agent Started Run: 2bp9e1u9
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/2bp9e1u9 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjJicDllMXU5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
13.334896691143513
The number of items in train is:
27
The loss for epoch 0
0.4938850626349449
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/media/graph/graph_0_summary_eab20644.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/media
The running loss is: 15.006994970142841 The number of items in train is: 27 The loss for epoch 1 0.5558146285238089 1 The running loss is: 11.964153863489628 The number of items in train is: 27 The loss for epoch 2 0.44311680975887513
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-summary.json
The running loss is: 9.46366179548204 The number of items in train is: 27 The loss for epoch 3 0.3505059924252607 1 The running loss is: 7.935978587716818 The number of items in train is: 27 The loss for epoch 4 0.29392513287840066 The running loss is: 7.160205790773034 The number of items in train is: 27 The loss for epoch 5 0.26519280706566795 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-summary.json
The running loss is: 7.432532018050551 The number of items in train is: 27 The loss for epoch 6 0.2752789636315019 2 The running loss is: 6.930170862004161 The number of items in train is: 27 The loss for epoch 7 0.256672994889043 The running loss is: 7.583688434679061 The number of items in train is: 27 The loss for epoch 8 0.2808773494325578 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-summary.json
The running loss is: 6.852206656709313 The number of items in train is: 27 The loss for epoch 9 0.2537854317299746 Data saved to: 28_May_202005_58PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_58PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['2bp9e1u9'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 792.838745
66 2020-04-22 sub_region ... 66 839.469238
67 2020-04-23 sub_region ... 67 793.361755
68 2020-04-24 sub_region ... 68 806.429504
69 2020-04-25 sub_region ... 69 816.609619
70 2020-04-26 sub_region ... 70 828.707764
71 2020-04-27 sub_region ... 71 823.823608
72 2020-04-28 sub_region ... 72 829.279419
73 2020-04-29 sub_region ... 73 820.153503
74 2020-04-30 sub_region ... 74 806.989746
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/media/plotly/test_plot_20_4bc95361.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 2bp9e1u9
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175822-2bp9e1u9/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 2bp9e1u9 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: izbxufpu with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 3 use_mask: True wandb: Agent Started Run: izbxufpu
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/izbxufpu INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOml6Ynh1ZnB1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
18.989742126315832
The number of items in train is:
26
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/config.yaml
The loss for epoch 0 0.7303746971659936
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/media/graph/graph_0_summary_0b88f687.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/media/graph
The running loss is: 16.38419210910797 The number of items in train is: 26 The loss for epoch 1 0.6301612349656912 1 The running loss is: 15.279926925897598 The number of items in train is: 26 The loss for epoch 2 0.5876894971499076
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-history.jsonl
The running loss is: 10.945733822882175 The number of items in train is: 26 The loss for epoch 3 0.4209897624185452 1 The running loss is: 5.606289468705654 The number of items in train is: 26 The loss for epoch 4 0.21562651802714056 2 The running loss is: 4.256948810070753 The number of items in train is: 26 The loss for epoch 5 0.16372880038733667
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-history.jsonl
The running loss is: 4.631639575585723 The number of items in train is: 26 The loss for epoch 6 0.17813998367637396 1 The running loss is: 6.9152334509417415 The number of items in train is: 26 The loss for epoch 7 0.26597051734391314 The running loss is: 3.967709585092962 The number of items in train is: 26 The loss for epoch 8 0.15260421481126776
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-history.jsonl
The running loss is: 3.5715523255057633 The number of items in train is: 26 The loss for epoch 9 0.13736739713483706 Data saved to: 28_May_202005_58PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['izbxufpu']
Data saved to: 28_May_202005_58PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 814.696899
66 2020-04-22 sub_region ... 66 825.249207
67 2020-04-23 sub_region ... 67 843.944824
68 2020-04-24 sub_region ... 68 816.804077
69 2020-04-25 sub_region ... 69 796.358887
70 2020-04-26 sub_region ... 70 799.779358
71 2020-04-27 sub_region ... 71 801.697021
72 2020-04-28 sub_region ... 72 800.930298
73 2020-04-29 sub_region ... 73 815.445740
74 2020-04-30 sub_region ... 74 822.305115
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/media/plotly/test_plot_20_d82f34c9.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: izbxufpu
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175839-izbxufpu/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: izbxufpu INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: uuquf6mj with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 3 use_mask: False wandb: Agent Started Run: uuquf6mj
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/uuquf6mj INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnV1cXVmNm1qOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
18.905425552278757
The number of items in train is:
26
The loss for epoch 0
0.7271317520107214
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/media/graph/graph_0_summary_b0526367.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/media/graph
The running loss is: 16.324342384934425 The number of items in train is: 26 The loss for epoch 1 0.6278593224974779 1 The running loss is: 15.281524695456028 The number of items in train is: 26 The loss for epoch 2 0.5877509498252318
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-summary.json
The running loss is: 10.686566263437271 The number of items in train is: 26 The loss for epoch 3 0.41102177936297196 1 The running loss is: 6.059355471283197 The number of items in train is: 26 The loss for epoch 4 0.2330521335108922 The running loss is: 4.697019984945655 The number of items in train is: 26 The loss for epoch 5 0.1806546148056021
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-summary.json
The running loss is: 6.977002476342022 The number of items in train is: 26 The loss for epoch 6 0.26834624909007776 1 The running loss is: 5.771942357998341 The number of items in train is: 26 The loss for epoch 7 0.2219977829999362 The running loss is: 4.5645317947492 The number of items in train is: 26 The loss for epoch 8 0.17555891518266156
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-summary.json
The running loss is: 5.686972144991159 The number of items in train is: 26 The loss for epoch 9 0.21872969788427538 1 Data saved to: 28_May_202005_58PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_58PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['uuquf6mj'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 736.851746
66 2020-04-22 sub_region ... 66 772.893066
67 2020-04-23 sub_region ... 67 762.027588
68 2020-04-24 sub_region ... 68 752.332520
69 2020-04-25 sub_region ... 69 736.720947
70 2020-04-26 sub_region ... 70 741.464417
71 2020-04-27 sub_region ... 71 733.462769
72 2020-04-28 sub_region ... 72 760.639526
73 2020-04-29 sub_region ... 73 746.799805
74 2020-04-30 sub_region ... 74 770.807861
[21 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/media/plotly/test_plot_20_556d85cf.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: uuquf6mj
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175851-uuquf6mj/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: uuquf6mj INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: gbl3pckk with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 4 use_mask: True wandb: Agent Started Run: gbl3pckk
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/gbl3pckk INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmdibDNwY2trOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
16.948153391480446
The number of items in train is:
26
The loss for epoch 0
0.6518520535184786
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/media/graph/graph_0_summary_cf8f90f5.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/media
The running loss is: 13.85379159078002 The number of items in train is: 26 The loss for epoch 1 0.5328381381069238 The running loss is: 9.64835274219513 The number of items in train is: 26 The loss for epoch 2 0.37109049008442807 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-summary.json
The running loss is: 8.4732926171273 The number of items in train is: 26 The loss for epoch 3 0.3258958698895115 The running loss is: 5.7244476936757565 The number of items in train is: 26 The loss for epoch 4 0.22017106514137524 The running loss is: 6.5513798370957375 The number of items in train is: 26 The loss for epoch 5 0.25197614758060527 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-summary.json
The running loss is: 8.37942417897284 The number of items in train is: 26 The loss for epoch 6 0.3222855453451092 The running loss is: 4.463485406711698 The number of items in train is: 26 The loss for epoch 7 0.1716725156427576 1 The running loss is: 4.722438246943057 The number of items in train is: 26 The loss for epoch 8 0.18163224026704064
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-summary.json
The running loss is: 4.340588949620724 The number of items in train is: 26 The loss for epoch 9 0.1669457288315663 1 Data saved to: 28_May_202005_59PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_59PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['gbl3pckk']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 851.006592
66 2020-04-22 sub_region ... 66 861.678711
67 2020-04-23 sub_region ... 67 870.574524
68 2020-04-24 sub_region ... 68 854.424133
69 2020-04-25 sub_region ... 69 835.877502
70 2020-04-26 sub_region ... 70 840.583374
71 2020-04-27 sub_region ... 71 839.167969
72 2020-04-28 sub_region ... 72 850.335999
73 2020-04-29 sub_region ... 73 848.938354
74 2020-04-30 sub_region ... 74 859.935547
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/media/plotly/test_plot_20_0f4d1d99.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: gbl3pckk
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175902-gbl3pckk/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: gbl3pckk INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: blhbm0pa with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 4 use_mask: False wandb: Agent Started Run: blhbm0pa
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/blhbm0pa INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmJsaGJtMHBhOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
16.634709760546684
The number of items in train is:
26
The loss for epoch 0
0.6397965292517955
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/media/graph/graph_0_summary_ce932c35.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/media
The running loss is: 14.594469293951988 The number of items in train is: 26 The loss for epoch 1 0.5613257420750765 The running loss is: 9.017393708229065 The number of items in train is: 26 The loss for epoch 2 0.3468228349318871 1 The running loss is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-summary.json
7.2338755913078785
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-history.jsonl
The number of items in train is: 26 The loss for epoch 3 0.27822598428107226 The running loss is: 4.978128891438246 The number of items in train is: 26 The loss for epoch 4 0.19146649582454792 1 The running loss is: 5.498803533613682 The number of items in train is: 26 The loss for epoch 5 0.21149244360052621 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-history.jsonl
The running loss is: 7.14033218100667 The number of items in train is: 26 The loss for epoch 6 0.27462816080794883 3 Stopping model now Data saved to: 28_May_202005_59PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_59PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 736.526611
66 2020-04-22 sub_region ... 66 758.064575
67 2020-04-23 sub_region ... 67 751.773193
68 2020-04-24 sub_region ... 68 746.766174
69 2020-04-25 sub_region ... 69 738.920166
70 2020-04-26 sub_region ... 70 743.997314
71 2020-04-27 sub_region ... 71 741.321045
72 2020-04-28 sub_region ... 72 750.091064
73 2020-04-29 sub_region ... 73 743.901367
74 2020-04-30 sub_region ... 74 754.618958
[21 rows x 28 columns]
INFO:wandb.wandb_agent:Running runs: ['blhbm0pa']
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/media/plotly/test_plot_14_251cab71.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: blhbm0pa
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/media/plotly/test_plot_all_15_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175914-blhbm0pa/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: blhbm0pa INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 9lg5on60 with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 5 use_mask: True wandb: Agent Started Run: 9lg5on60
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/9lg5on60 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjlsZzVvbjYwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
14.528097435832024
The number of items in train is:
25
The loss for epoch 0
0.581123897433281
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/media/graph/graph_0_summary_fa07a0d1.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/media
The running loss is: 14.321749061346054 The number of items in train is: 25 The loss for epoch 1 0.5728699624538421 The running loss is: 8.53690830618143 The number of items in train is: 25 The loss for epoch 2 0.3414763322472572 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-summary.json
The running loss is: 3.877267986536026 The number of items in train is: 25 The loss for epoch 3 0.15509071946144104 The running loss is: 3.9804319692775607 The number of items in train is: 25 The loss for epoch 4 0.15921727877110242 1 The running loss is: 4.088011069223285 The number of items in train is: 25 The loss for epoch 5 0.16352044276893138
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-summary.json
The running loss is: 3.642200954724103 The number of items in train is: 25 The loss for epoch 6 0.14568803818896414 1 The running loss is: 4.692892727442086 The number of items in train is: 25 The loss for epoch 7 0.18771570909768343 2 The running loss is: 3.2474756240844727 The number of items in train is: 25 The loss for epoch 8 0.1298990249633789
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-summary.json
The running loss is: 3.19531544810161 The number of items in train is: 25 The loss for epoch 9 0.1278126179240644 Data saved to: 28_May_202005_59PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_59PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['9lg5on60'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 711.413086
66 2020-04-22 sub_region ... 66 714.489502
67 2020-04-23 sub_region ... 67 749.420166
68 2020-04-24 sub_region ... 68 730.427063
69 2020-04-25 sub_region ... 69 707.012146
70 2020-04-26 sub_region ... 70 700.846252
71 2020-04-27 sub_region ... 71 701.634644
72 2020-04-28 sub_region ... 72 702.439697
73 2020-04-29 sub_region ... 73 710.740356
74 2020-04-30 sub_region ... 74 736.382812
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/media/plotly/test_plot_20_b4b65440.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 9lg5on60
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175925-9lg5on60/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 9lg5on60 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: l04ebxec with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 5 use_mask: False wandb: Agent Started Run: l04ebxec
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/l04ebxec INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmwwNGVieGVjOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
14.583468616008759
The number of items in train is:
25
The loss for epoch 0
0.5833387446403503
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/media/graph/graph_0_summary_099d89b2.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/media
The running loss is: 11.234198205173016 The number of items in train is: 25 The loss for epoch 1 0.4493679282069206 1 The running loss is: 6.8342317044734955 The number of items in train is: 25 The loss for epoch 2 0.2733692681789398 2 The running loss is: 4.914771635085344 The number of items in train is: 25 The loss for epoch 3 0.19659086540341378
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-summary.json
The running loss is: 6.097113943658769 The number of items in train is: 25 The loss for epoch 4 0.24388455774635076 1 The running loss is: 5.855831284075975 The number of items in train is: 25 The loss for epoch 5 0.23423325136303902 2 The running loss is: 4.823610462248325 The number of items in train is: 25 The loss for epoch 6 0.19294441848993302
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-summary.json
The running loss is: 5.116255223751068 The number of items in train is: 25 The loss for epoch 7 0.2046502089500427 1 The running loss is: 3.9354080138728023 The number of items in train is: 25 The loss for epoch 8 0.1574163205549121 2 The running loss is: 3.635712749324739 The number of items in train is: 25 The loss for epoch 9 0.14542850997298956
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-summary.json
3 Stopping model now Data saved to: 28_May_202005_59PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202005_59PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['l04ebxec']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-summary.json
date level ... original_index preds 54 2020-04-10 sub_region ... 54 0.000000 55 2020-04-11 sub_region ... 55 0.000000 56 2020-04-12 sub_region ... 56 0.000000 57 2020-04-13 sub_region ... 57 0.000000 58 2020-04-14 sub_region ... 58 0.000000 59 2020-04-15 sub_region ... 59 0.000000 60 2020-04-16 sub_region ... 60 0.000000 61 2020-04-17 sub_region ... 61 0.000000 62 2020-04-18 sub_region ... 62 0.000000 63 2020-04-19 sub_region ... 63 0.000000 64 2020-04-20 sub_region ... 64 0.000000 65 2020-04-21 sub_region ... 65 742.793396 66 2020-04-22 sub_region ... 66 736.652344 67 2020-04-23 sub_region ... 67 754.830688 68 2020-04-24 sub_region ... 68 749.173950 69 2020-04-25 sub_region ... 69 738.829346 70 2020-04-26 sub_region ... 70 734.617004 71 2020-04-27 sub_region ... 71 732.794556 72 2020-04-28 sub_region ... 72 736.974976 73 2020-04-29 sub_region ... 73 737.531616 74 2020-04-30 sub_region ... 74 753.737549 [21 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/media/plotly/test_plot_20_71421788.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: l04ebxec
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175942-l04ebxec/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: l04ebxec INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: sp9yc18q with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 1 use_mask: True wandb: Agent Started Run: sp9yc18q
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/sp9yc18q INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnNwOXljMThxOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/media/graph/graph_0_summary_c09f855e.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/media
The running loss is: 12.349453534625354 The number of items in train is: 27 The loss for epoch 0 0.4573871679490872 The running loss is: 30.535183822968975 The number of items in train is: 27 The loss for epoch 1 1.130932734184036
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/wandb-summary.json
The running loss is: 18.304419979453087 The number of items in train is: 27 The loss for epoch 2 0.6779414807204847 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/wandb-summary.json
The running loss is: 10.870923589682207 The number of items in train is: 27 The loss for epoch 3 0.4026267996178595 2 The running loss is: 9.898105231113732 The number of items in train is: 27 The loss for epoch 4 0.36659649004124933 3 Stopping model now Data saved to: 28_May_202006_00PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/wandb-summary.json
Data saved to: 28_May_202006_00PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 INFO:wandb.wandb_agent:Running runs: ['sp9yc18q'] /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 502.529541
66 2020-04-22 sub_region ... 66 492.594849
67 2020-04-23 sub_region ... 67 496.524628
68 2020-04-24 sub_region ... 68 495.569824
69 2020-04-25 sub_region ... 69 495.434570
70 2020-04-26 sub_region ... 70 493.173859
71 2020-04-27 sub_region ... 71 493.890808
72 2020-04-28 sub_region ... 72 491.183136
73 2020-04-29 sub_region ... 73 497.434570
74 2020-04-30 sub_region ... 74 489.901245
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: sp9yc18q
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/media/plotly/test_plot_10_c557f1bd.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/media/plotly/test_plot_all_11_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_175959-sp9yc18q/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: sp9yc18q INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: rf529106 with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 1 use_mask: False wandb: Agent Started Run: rf529106
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/rf529106 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnJmNTI5MTA2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/media/graph/graph_0_summary_0a959aa4.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/media/graph
The running loss is: 12.00125639885664 The number of items in train is: 27 The loss for epoch 0 0.4444909777354311 The running loss is: 27.564760353183374 The number of items in train is: 27 The loss for epoch 1 1.0209170501179028
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-history.jsonl
The running loss is: 19.1493139838567 The number of items in train is: 27 The loss for epoch 2 0.7092338512539519 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-history.jsonl
The running loss is: 9.482985021779314 The number of items in train is: 27 The loss for epoch 3 0.3512216674733079 The running loss is: 8.379846808500588 The number of items in train is: 27 The loss for epoch 4 0.31036469661113286 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-history.jsonl
The running loss is: 6.564052293615532 The number of items in train is: 27 The loss for epoch 5 0.24311304791168636
INFO:wandb.wandb_agent:Running runs: ['rf529106']
The running loss is: 7.043116731874761 The number of items in train is: 27 The loss for epoch 6 0.2608561752546208
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-history.jsonl
The running loss is: 6.5104550529504195 The number of items in train is: 27 The loss for epoch 7 0.2411279649240896 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-history.jsonl
The running loss is: 7.045392430853099 The number of items in train is: 27 The loss for epoch 8 0.2609404604019666 The running loss is: 6.926817309111357 The number of items in train is: 27 The loss for epoch 9 0.25654878922634655 1 Data saved to: 28_May_202006_00PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-history.jsonl
Data saved to: 28_May_202006_00PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/config.yaml /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 778.043945
66 2020-04-22 sub_region ... 66 783.455933
67 2020-04-23 sub_region ... 67 766.487671
68 2020-04-24 sub_region ... 68 773.625916
69 2020-04-25 sub_region ... 69 782.676147
70 2020-04-26 sub_region ... 70 784.909546
71 2020-04-27 sub_region ... 71 784.748901
72 2020-04-28 sub_region ... 72 778.614258
73 2020-04-29 sub_region ... 73 782.117676
74 2020-04-30 sub_region ... 74 761.960938
[21 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/media/plotly/test_plot_20_647b6c30.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: rf529106
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180011-rf529106/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: rf529106 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: eupck2gp with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 2 use_mask: True wandb: Agent Started Run: eupck2gp
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/eupck2gp INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmV1cGNrMmdwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/media/graph/graph_0_summary_0ad3907c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/media/graph
The running loss is: 15.362042212858796 The number of items in train is: 27 The loss for epoch 0 0.5689645264021777 The running loss is: 18.604257106781006 The number of items in train is: 27 The loss for epoch 1 0.6890465595104076 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-summary.json
The running loss is: 19.387704892084002 The number of items in train is: 27 The loss for epoch 2 0.7180631441512594
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-summary.json
The running loss is: 17.115502156317234 The number of items in train is: 27 The loss for epoch 3 0.6339074872710087 The running loss is: 9.399747041985393 The number of items in train is: 27 The loss for epoch 4 0.3481387793327923 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-summary.json
The running loss is: 8.062234118580818 The number of items in train is: 27 The loss for epoch 5 0.2986012636511414
INFO:wandb.wandb_agent:Running runs: ['eupck2gp']
The running loss is: 7.508644045330584 The number of items in train is: 27 The loss for epoch 6 0.2780979276048364
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-summary.json
The running loss is: 7.624856340698898 The number of items in train is: 27 The loss for epoch 7 0.2824020866925518 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-summary.json
The running loss is: 7.4580579325556755 The number of items in train is: 27 The loss for epoch 8 0.27622436787243243 The running loss is: 6.958448266610503 The number of items in train is: 27 The loss for epoch 9 0.25772030617075936 Data saved to: 28_May_202006_00PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-summary.json
Data saved to: 28_May_202006_00PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 772.555298
66 2020-04-22 sub_region ... 66 766.406433
67 2020-04-23 sub_region ... 67 773.479492
68 2020-04-24 sub_region ... 68 770.044922
69 2020-04-25 sub_region ... 69 772.980347
70 2020-04-26 sub_region ... 70 772.595215
71 2020-04-27 sub_region ... 71 772.935425
72 2020-04-28 sub_region ... 72 770.757202
73 2020-04-29 sub_region ... 73 772.154602
74 2020-04-30 sub_region ... 74 771.334900
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/media/plotly/test_plot_20_cf3ed598.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: eupck2gp
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180028-eupck2gp/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: eupck2gp INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: o5ynrlxu with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 2 use_mask: False wandb: Agent Started Run: o5ynrlxu
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/o5ynrlxu INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm81eW5ybHh1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/media/graph/graph_0_summary_44fabac8.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/media
The running loss is: 15.227613762952387 The number of items in train is: 27 The loss for epoch 0 0.5639856949241625 The running loss is: 17.898133425042033 The number of items in train is: 27 The loss for epoch 1 0.6628938305571124
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-summary.json
1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-history.jsonl
The running loss is: 15.654380347579718 The number of items in train is: 27 The loss for epoch 2 0.5797918647251747
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-history.jsonl
The running loss is: 9.89348728954792 The number of items in train is: 27 The loss for epoch 3 0.3664254551684415 1 The running loss is: 8.431944162817672 The number of items in train is: 27 The loss for epoch 4 0.31229422825250636
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-history.jsonl
The running loss is: 8.609166414942592 The number of items in train is: 27 The loss for epoch 5 0.31885801536824415 1
INFO:wandb.wandb_agent:Running runs: ['o5ynrlxu'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-history.jsonl
The running loss is: 9.016866168007255 The number of items in train is: 27 The loss for epoch 6 0.3339580062224909 2 The running loss is: 8.0558643322438 The number of items in train is: 27 The loss for epoch 7 0.29836534563865924
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-history.jsonl
The running loss is: 8.685117572546005 The number of items in train is: 27 The loss for epoch 8 0.3216710212054076 1 The running loss is: 6.745234100148082 The number of items in train is: 27 The loss for epoch 9 0.2498234851906697 Data saved to: 28_May_202006_00PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-history.jsonl
Data saved to: 28_May_202006_00PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 783.742432
66 2020-04-22 sub_region ... 66 782.695801
67 2020-04-23 sub_region ... 67 783.955322
68 2020-04-24 sub_region ... 68 783.325439
69 2020-04-25 sub_region ... 69 783.187744
70 2020-04-26 sub_region ... 70 783.111938
71 2020-04-27 sub_region ... 71 783.345886
72 2020-04-28 sub_region ... 72 782.962280
73 2020-04-29 sub_region ... 73 782.751831
74 2020-04-30 sub_region ... 74 783.127808
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/media/plotly/test_plot_20_86f33ef2.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: o5ynrlxu
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180044-o5ynrlxu/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: o5ynrlxu INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: c6rdxmat with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 3 use_mask: True wandb: Agent Started Run: c6rdxmat
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/c6rdxmat INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmM2cmR4bWF0OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/media/graph/graph_0_summary_bf741941.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/media/graph
The running loss is: 21.012039236724377 The number of items in train is: 26 The loss for epoch 0 0.8081553552586299 The running loss is: 16.833910040557384 The number of items in train is: 26 The loss for epoch 1 0.6474580784829763 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-summary.json
The running loss is: 16.23337048664689 The number of items in train is: 26 The loss for epoch 2 0.6243604033325727
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-summary.json
The running loss is: 15.538650223985314 The number of items in train is: 26 The loss for epoch 3 0.5976403932302043 1 The running loss is: 15.044064596295357 The number of items in train is: 26 The loss for epoch 4 0.5786178690882829 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-summary.json
The running loss is: 11.831363804638386 The number of items in train is: 26 The loss for epoch 5 0.4550524540245533
INFO:wandb.wandb_agent:Running runs: ['c6rdxmat']
The running loss is: 7.703582746908069 The number of items in train is: 26 The loss for epoch 6 0.2962916441118488 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-summary.json
The running loss is: 8.853043738752604 The number of items in train is: 26 The loss for epoch 7 0.3405016822597155 2 The running loss is: 5.49346605874598 The number of items in train is: 26 The loss for epoch 8 0.21128715610561463
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-summary.json
The running loss is: 6.049623145721853 The number of items in train is: 26 The loss for epoch 9 0.23267781329699433 Data saved to: 28_May_202006_01PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_01PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 793.348389
66 2020-04-22 sub_region ... 66 793.042908
67 2020-04-23 sub_region ... 67 790.546631
68 2020-04-24 sub_region ... 68 791.214233
69 2020-04-25 sub_region ... 69 792.006836
70 2020-04-26 sub_region ... 70 792.608032
71 2020-04-27 sub_region ... 71 792.963989
72 2020-04-28 sub_region ... 72 789.634521
73 2020-04-29 sub_region ... 73 793.287476
74 2020-04-30 sub_region ... 74 785.882202
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/media/plotly/test_plot_20_ef160644.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: c6rdxmat
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180101-c6rdxmat/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: c6rdxmat INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: nnrubgkf with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 3 use_mask: False wandb: Agent Started Run: nnrubgkf
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/nnrubgkf INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm5ucnViZ2tmOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/media/graph/graph_0_summary_c4d34699.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/media
The running loss is: 20.714700505137444 The number of items in train is: 26 The loss for epoch 0 0.7967192501975939 The running loss is: 16.891354955732822 The number of items in train is: 26 The loss for epoch 1 0.6496674982974162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/wandb-summary.json
The running loss is: 16.139072712510824 The number of items in train is: 26 The loss for epoch 2 0.620733565865801 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/wandb-summary.json
The running loss is: 15.367559241130948 The number of items in train is: 26 The loss for epoch 3 0.5910599708127288 2 The running loss is: 14.148938469588757 The number of items in train is: 26 The loss for epoch 4 0.5441899411380291 3 Stopping model now Data saved to: 28_May_202006_01PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/wandb-summary.json
Data saved to: 28_May_202006_01PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
INFO:wandb.wandb_agent:Running runs: ['nnrubgkf'] /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 440.134735
66 2020-04-22 sub_region ... 66 439.800903
67 2020-04-23 sub_region ... 67 443.283325
68 2020-04-24 sub_region ... 68 442.842438
69 2020-04-25 sub_region ... 69 440.653381
70 2020-04-26 sub_region ... 70 441.242767
71 2020-04-27 sub_region ... 71 440.577118
72 2020-04-28 sub_region ... 72 442.682800
73 2020-04-29 sub_region ... 73 441.382904
74 2020-04-30 sub_region ... 74 445.453522
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: nnrubgkf
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/media/plotly/test_plot_10_0d49373a.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/media/plotly/test_plot_all_11_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180118-nnrubgkf/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: nnrubgkf INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: auxmbkll with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 4 use_mask: True wandb: Agent Started Run: auxmbkll
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/auxmbkll INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmF1eG1ia2xsOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/media/graph/graph_0_summary_61198c9f.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/media/graph
The running loss is: 23.482627481222153 The number of items in train is: 26 The loss for epoch 0 0.9031779800470059 The running loss is: 20.23513476550579 The number of items in train is: 26 The loss for epoch 1 0.778274414057915 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-history.jsonl
The running loss is: 21.649361565709114 The number of items in train is: 26 The loss for epoch 2 0.8326677525272737
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-history.jsonl
The running loss is: 15.995433270931244 The number of items in train is: 26 The loss for epoch 3 0.615208971958894 1 The running loss is: 12.781472019851208 The number of items in train is: 26 The loss for epoch 4 0.4915950776865849 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-history.jsonl
The running loss is: 8.850842893123627 The number of items in train is: 26 The loss for epoch 5 0.3404170343509087
INFO:wandb.wandb_agent:Running runs: ['auxmbkll']
The running loss is: 6.8902468085289 The number of items in train is: 26 The loss for epoch 6 0.26500949263572693 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-history.jsonl
The running loss is: 8.579151637852192 The number of items in train is: 26 The loss for epoch 7 0.32996737068662274 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-history.jsonl
The running loss is: 6.014725340530276 The number of items in train is: 26 The loss for epoch 8 0.23133559002039525 The running loss is: 4.920796798542142 The number of items in train is: 26 The loss for epoch 9 0.18926141532854393 1 Data saved to: 28_May_202006_01PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-history.jsonl
Data saved to: 28_May_202006_01PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 784.844604
66 2020-04-22 sub_region ... 66 785.061829
67 2020-04-23 sub_region ... 67 785.951660
68 2020-04-24 sub_region ... 68 785.422729
69 2020-04-25 sub_region ... 69 784.747925
70 2020-04-26 sub_region ... 70 784.533691
71 2020-04-27 sub_region ... 71 784.408447
72 2020-04-28 sub_region ... 72 785.171936
73 2020-04-29 sub_region ... 73 784.649170
74 2020-04-30 sub_region ... 74 785.943665
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: auxmbkll
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/media/plotly/test_plot_20_bf33d510.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180132-auxmbkll/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: auxmbkll INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: srhmp681 with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 4 use_mask: False wandb: Agent Started Run: srhmp681
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/srhmp681 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnNyaG1wNjgxOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/media/graph/graph_0_summary_dded512a.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/media/graph
The running loss is: 23.31422958523035 The number of items in train is: 26 The loss for epoch 0 0.896701137893475 The running loss is: 19.628748193383217 The number of items in train is: 26 The loss for epoch 1 0.7549518535916622 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-summary.json
The running loss is: 22.03840383887291 The number of items in train is: 26 The loss for epoch 2 0.8476309168797272 The running loss is: 15.539160393178463 The number of items in train is: 26
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-history.jsonl
The loss for epoch 3 0.5976600151222485
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-summary.json
1 The running loss is: 12.818177118897438 The number of items in train is: 26 The loss for epoch 4 0.4930068122652861
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-summary.json
The running loss is: 10.63734956085682 The number of items in train is: 26 The loss for epoch 5 0.4091288292637238 1 The running loss is: 6.949803344905376 The number of items in train is: 26
INFO:wandb.wandb_agent:Running runs: ['srhmp681']
The loss for epoch 6 0.2673001286502068
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-summary.json
The running loss is: 7.64347056671977 The number of items in train is: 26 The loss for epoch 7 0.29397963718152964 1 The running loss is: 11.507175385951996 The number of items in train is: 26 The loss for epoch 8 0.4425836686904614
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-summary.json
The running loss is: 5.667488571256399 The number of items in train is: 26 The loss for epoch 9 0.21798032966370767 Data saved to: 28_May_202006_01PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_01PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 749.537842
66 2020-04-22 sub_region ... 66 753.229736
67 2020-04-23 sub_region ... 67 753.515015
68 2020-04-24 sub_region ... 68 753.205444
69 2020-04-25 sub_region ... 69 751.122925
70 2020-04-26 sub_region ... 70 750.473755
71 2020-04-27 sub_region ... 71 748.162842
72 2020-04-28 sub_region ... 72 752.245361
73 2020-04-29 sub_region ... 73 750.474243
74 2020-04-30 sub_region ... 74 754.320679
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: srhmp681
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/media/plotly/test_plot_20_67e12e5e.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180149-srhmp681/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: srhmp681 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: n3ev68is with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 5 use_mask: True wandb: Agent Started Run: n3ev68is
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/n3ev68is INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm4zZXY2OGlzOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/media/graph/graph_0_summary_ab0d380c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/media
The running loss is: 17.830005079507828 The number of items in train is: 25 The loss for epoch 0 0.7132002031803131 The running loss is: 16.9639795422554 The number of items in train is: 25 The loss for epoch 1 0.6785591816902161 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-summary.json
The running loss is: 19.417406290769577 The number of items in train is: 25 The loss for epoch 2 0.776696251630783 The running loss is: 13.737709395587444 The number of items in train is: 25 The loss for epoch 3 0.5495083758234978
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-summary.json
1 The running loss is: 13.103170596063137 The number of items in train is: 25 The loss for epoch 4 0.5241268238425255 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-summary.json
The running loss is: 9.996175795793533 The number of items in train is: 25 The loss for epoch 5 0.3998470318317413 3 Stopping model now Data saved to: 28_May_202006_02PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_02PM_model.pth interpolate should be below
INFO:wandb.wandb_agent:Running runs: ['n3ev68is']
Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 417.858063
66 2020-04-22 sub_region ... 66 419.026245
67 2020-04-23 sub_region ... 67 420.017090
68 2020-04-24 sub_region ... 68 420.572571
69 2020-04-25 sub_region ... 69 419.899597
70 2020-04-26 sub_region ... 70 419.641113
71 2020-04-27 sub_region ... 71 419.378143
72 2020-04-28 sub_region ... 72 419.695251
73 2020-04-29 sub_region ... 73 419.175232
74 2020-04-30 sub_region ... 74 421.516815
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-history.jsonl
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/media/plotly/test_plot_12_d17e53a2.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: n3ev68is
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/media/plotly/test_plot_all_13_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180205-n3ev68is/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: n3ev68is INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: yrue8x18 with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 5 use_mask: False wandb: Agent Started Run: yrue8x18
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/yrue8x18 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnlydWU4eDE4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/media/graph/graph_0_summary_38b94582.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/media
The running loss is: 17.741086795926094 The number of items in train is: 25 The loss for epoch 0 0.7096434718370438 The running loss is: 16.730679839849472 The number of items in train is: 25 The loss for epoch 1 0.6692271935939789 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-summary.json
The running loss is: 19.939407482743263 The number of items in train is: 25 The loss for epoch 2 0.7975762993097305 The running loss is: 13.255669929087162 The number of items in train is: 25 The loss for epoch 3 0.5302267971634865
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-summary.json
1 The running loss is: 11.027616888284683 The number of items in train is: 25 The loss for epoch 4 0.4411046755313873 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-summary.json
The running loss is: 6.718288138508797 The number of items in train is: 25 The loss for epoch 5 0.26873152554035185 The running loss is: 7.313029149547219 The number of items in train is: 25 The loss for epoch 6 0.29252116598188876 1
INFO:wandb.wandb_agent:Running runs: ['yrue8x18'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-summary.json
The running loss is: 12.773601979017258 The number of items in train is: 25 The loss for epoch 7 0.5109440791606903 The running loss is: 6.80379575304687 The number of items in train is: 25 The loss for epoch 8 0.2721518301218748 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-summary.json
The running loss is: 4.739170776680112 The number of items in train is: 25 The loss for epoch 9 0.18956683106720448 2 Data saved to: 28_May_202006_02PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_02PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 770.853760
66 2020-04-22 sub_region ... 66 769.976440
67 2020-04-23 sub_region ... 67 771.220154
68 2020-04-24 sub_region ... 68 770.911987
69 2020-04-25 sub_region ... 69 770.517700
70 2020-04-26 sub_region ... 70 770.186768
71 2020-04-27 sub_region ... 71 770.193970
72 2020-04-28 sub_region ... 72 770.157593
73 2020-04-29 sub_region ... 73 770.232788
74 2020-04-30 sub_region ... 74 770.960083
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/media/plotly/test_plot_20_57ec849c.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: yrue8x18
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180217-yrue8x18/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: yrue8x18 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: s5c7gh6g with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 1 use_mask: True wandb: Agent Started Run: s5c7gh6g
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/s5c7gh6g INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnM1YzdnaDZnOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/media/graph/graph_0_summary_25258857.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/media/graph
The running loss is: 13.64892402663827 The number of items in train is: 27 The loss for epoch 0 0.5055157046903063
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-history.jsonl
The running loss is: 37.692751033231616 The number of items in train is: 27 The loss for epoch 1 1.3960278160456154
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-history.jsonl
The running loss is: 20.47567129507661 The number of items in train is: 27 The loss for epoch 2 0.7583581961139485
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-history.jsonl
The running loss is: 20.933722829446197 The number of items in train is: 27 The loss for epoch 3 0.7753230677572666 1
INFO:wandb.wandb_agent:Running runs: ['s5c7gh6g'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-history.jsonl
The running loss is: 20.798208009451628 The number of items in train is: 27 The loss for epoch 4 0.7703040003500603 The running loss is: 20.184429235756397 The number of items in train is: 27 The loss for epoch 5 0.7475714531761629
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-history.jsonl
1 The running loss is: 20.411993622779846 The number of items in train is: 27 The loss for epoch 6 0.755999763806661 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-history.jsonl
The running loss is: 19.94804622977972 The number of items in train is: 27 The loss for epoch 7 0.7388165270288786 3 Stopping model now Data saved to: 28_May_202006_02PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-history.jsonl
Data saved to: 28_May_202006_02PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 474.087341
66 2020-04-22 sub_region ... 66 474.107361
67 2020-04-23 sub_region ... 67 473.755005
68 2020-04-24 sub_region ... 68 474.003113
69 2020-04-25 sub_region ... 69 474.199402
70 2020-04-26 sub_region ... 70 474.221405
71 2020-04-27 sub_region ... 71 474.191833
72 2020-04-28 sub_region ... 72 473.996887
73 2020-04-29 sub_region ... 73 474.185181
74 2020-04-30 sub_region ... 74 473.697998
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/media/plotly/test_plot_16_1f8d7214.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: s5c7gh6g
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/media/plotly/test_plot_all_17_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180233-s5c7gh6g/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: s5c7gh6g INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: zt8tb192 with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 1 use_mask: False wandb: Agent Started Run: zt8tb192
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/zt8tb192 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnp0OHRiMTkyOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/media/graph/graph_0_summary_7ab4a493.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/media/graph
The running loss is: 13.37478191871196 The number of items in train is: 27 The loss for epoch 0 0.49536229328562814
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-summary.json
The running loss is: 37.45775743201375 The number of items in train is: 27 The loss for epoch 1 1.3873243493338425
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-summary.json
The running loss is: 20.595209177583456 The number of items in train is: 27 The loss for epoch 2 0.7627855250956835
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-summary.json
The running loss is: 21.006036780774593 The number of items in train is: 27 The loss for epoch 3 0.7780013622509109
INFO:wandb.wandb_agent:Running runs: ['zt8tb192'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-summary.json
The running loss is: 20.73020259477198 The number of items in train is: 27 The loss for epoch 4 0.7677852812878512
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-summary.json
The running loss is: 20.24301341921091 The number of items in train is: 27 The loss for epoch 5 0.7497412377485523 1 The running loss is: 20.167157787829638 The number of items in train is: 27 The loss for epoch 6 0.7469317699196162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-summary.json
2 The running loss is: 18.895125970244408 The number of items in train is: 27 The loss for epoch 7 0.6998194803794225 3 Stopping model now Data saved to: 28_May_202006_02PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-summary.json
Data saved to: 28_May_202006_02PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 474.395355
66 2020-04-22 sub_region ... 66 474.474243
67 2020-04-23 sub_region ... 67 473.946503
68 2020-04-24 sub_region ... 68 474.251740
69 2020-04-25 sub_region ... 69 474.498688
70 2020-04-26 sub_region ... 70 474.531128
71 2020-04-27 sub_region ... 71 474.499969
72 2020-04-28 sub_region ... 72 474.257050
73 2020-04-29 sub_region ... 73 474.527222
74 2020-04-30 sub_region ... 74 473.818420
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/media/plotly/test_plot_16_c257944a.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: zt8tb192
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/media/plotly/test_plot_all_17_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180250-zt8tb192/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: zt8tb192 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: knbqwf8h with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 2 use_mask: True wandb: Agent Started Run: knbqwf8h
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/knbqwf8h INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmtuYnF3ZjhoOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/media/graph/graph_0_summary_e54907c4.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/media
The running loss is: 15.137208757922053 The number of items in train is: 27 The loss for epoch 0 0.5606373614045205
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-summary.json
The running loss is: 30.88287841156125 The number of items in train is: 27 The loss for epoch 1 1.1438103115393057
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-summary.json
The running loss is: 25.371903955936432 The number of items in train is: 27 The loss for epoch 2 0.9397001465161642
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-summary.json
The running loss is: 20.891498629003763 The number of items in train is: 27 The loss for epoch 3 0.7737592084816208
INFO:wandb.wandb_agent:Running runs: ['knbqwf8h'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-summary.json
The running loss is: 20.58536821603775 The number of items in train is: 27 The loss for epoch 4 0.7624210450384352 1 The running loss is: 21.16692252457142 The number of items in train is: 27 The loss for epoch 5 0.7839600935026452
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-summary.json
2 The running loss is: 20.775313809514046 The number of items in train is: 27 The loss for epoch 6 0.7694560670190387 3 Stopping model now Data saved to: 28_May_202006_03PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-summary.json
Data saved to: 28_May_202006_03PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 422.063049
66 2020-04-22 sub_region ... 66 422.065277
67 2020-04-23 sub_region ... 67 422.008148
68 2020-04-24 sub_region ... 68 422.044128
69 2020-04-25 sub_region ... 69 422.075989
70 2020-04-26 sub_region ... 70 422.085114
71 2020-04-27 sub_region ... 71 422.084900
72 2020-04-28 sub_region ... 72 422.078888
73 2020-04-29 sub_region ... 73 422.060028
74 2020-04-30 sub_region ... 74 422.037537
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: knbqwf8h
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/media/plotly/test_plot_all_15_6655637c.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/media/plotly/test_plot_14_7a77c25e.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180312-knbqwf8h/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: knbqwf8h INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: jmxt5fwy with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 2 use_mask: False wandb: Agent Started Run: jmxt5fwy
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/jmxt5fwy INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-metadata.json INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmpteHQ1Znd5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/media/graph/graph_0_summary_2d11b813.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/media/graph
The running loss is: 15.071377269923687 The number of items in train is: 27 The loss for epoch 0 0.5581991581453217
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-history.jsonl
The running loss is: 31.568736284971237 The number of items in train is: 27 The loss for epoch 1 1.1692124549989347
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-history.jsonl
The running loss is: 25.198651179671288 The number of items in train is: 27 The loss for epoch 2 0.9332833770248625
INFO:wandb.wandb_agent:Running runs: ['jmxt5fwy'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-history.jsonl
The running loss is: 21.102718703448772 The number of items in train is: 27 The loss for epoch 3 0.7815821742018064
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-history.jsonl
The running loss is: 20.57193885743618 The number of items in train is: 27 The loss for epoch 4 0.7619236613865252 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-history.jsonl
The running loss is: 21.114420972764492 The number of items in train is: 27 The loss for epoch 5 0.7820155915838701 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-history.jsonl
The running loss is: 21.067522056400776 The number of items in train is: 27 The loss for epoch 6 0.7802785946815102 3 Stopping model now Data saved to: 28_May_202006_03PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_03PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 423.483063
66 2020-04-22 sub_region ... 66 423.511169
67 2020-04-23 sub_region ... 67 423.472900
68 2020-04-24 sub_region ... 68 423.498932
69 2020-04-25 sub_region ... 69 423.520294
70 2020-04-26 sub_region ... 70 423.524902
71 2020-04-27 sub_region ... 71 423.520996
72 2020-04-28 sub_region ... 72 423.510651
73 2020-04-29 sub_region ... 73 423.502411
74 2020-04-30 sub_region ... 74 423.487152
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/media/plotly/test_plot_14_0bb7c432.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: jmxt5fwy
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/media/plotly/test_plot_all_15_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180328-jmxt5fwy/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: jmxt5fwy INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: wd43ji3h with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 3 use_mask: True wandb: Agent Started Run: wd43ji3h
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/wd43ji3h INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOndkNDNqaTNoOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/media/graph/graph_0_summary_0d3b0650.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/media
The running loss is: 24.238217145204544 The number of items in train is: 26 The loss for epoch 0 0.9322391209694055
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-summary.json
The running loss is: 16.11776690930128 The number of items in train is: 26 The loss for epoch 1 0.6199141118962032 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-summary.json
The running loss is: 16.947837613523006 The number of items in train is: 26 The loss for epoch 2 0.6518399082124233 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-summary.json
The running loss is: 17.64614214003086 The number of items in train is: 26 The loss for epoch 3 0.6786977746165715
INFO:wandb.wandb_agent:Running runs: ['wd43ji3h'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-summary.json
The running loss is: 16.301386021077633 The number of items in train is: 26 The loss for epoch 4 0.6269763854260628 The running loss is: 16.042810605838895 The number of items in train is: 26 The loss for epoch 5 0.6170311771476498
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-summary.json
The running loss is: 15.744363751262426 The number of items in train is: 26 The loss for epoch 6 0.6055524519716318 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-summary.json
The running loss is: 16.016306111589074 The number of items in train is: 26 The loss for epoch 7 0.6160117735226567 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-summary.json
The running loss is: 15.813427444547415 The number of items in train is: 26 The loss for epoch 8 0.6082087478672082
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-summary.json
The running loss is: 16.24589281156659 The number of items in train is: 26 The loss for epoch 9 0.6248420312140996 Data saved to: 28_May_202006_03PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_03PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 509.934052
66 2020-04-22 sub_region ... 66 507.839417
67 2020-04-23 sub_region ... 67 509.408234
68 2020-04-24 sub_region ... 68 509.694031
69 2020-04-25 sub_region ... 69 509.575317
70 2020-04-26 sub_region ... 70 508.830750
71 2020-04-27 sub_region ... 71 509.065674
72 2020-04-28 sub_region ... 72 507.174622
73 2020-04-29 sub_region ... 73 509.588074
74 2020-04-30 sub_region ... 74 507.856110
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: wd43ji3h
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/media/plotly/test_plot_20_75179c3f.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180346-wd43ji3h/wandb-events.jsonl INFO:wandb.wandb_agent:Cleaning up finished run: wd43ji3h INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: bfg54kzp with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 3 use_mask: False wandb: Agent Started Run: bfg54kzp
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/bfg54kzp INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmJmZzU0a3pwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/media/graph/graph_0_summary_c8daf13d.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/media
The running loss is: 24.46081606298685 The number of items in train is: 26 The loss for epoch 0 0.9408006178071866
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-history.jsonl
The running loss is: 16.266936726868153 The number of items in train is: 26 The loss for epoch 1 0.625651412571852 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-history.jsonl
The running loss is: 16.894336327910423 The number of items in train is: 26 The loss for epoch 2 0.6497821664580932
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-history.jsonl
The running loss is: 17.256735399365425 The number of items in train is: 26 The loss for epoch 3 0.6637205922832856
INFO:wandb.wandb_agent:Running runs: ['bfg54kzp'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-history.jsonl
The running loss is: 16.231802832335234 The number of items in train is: 26 The loss for epoch 4 0.6243001089359705 The running loss is: 16.1003286074847 The number of items in train is: 26 The loss for epoch 5 0.6192434079801807
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-history.jsonl
The running loss is: 15.810061607509851 The number of items in train is: 26 The loss for epoch 6 0.6080792925965327 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-history.jsonl
The running loss is: 16.048187194392085 The number of items in train is: 26 The loss for epoch 7 0.6172379690150802 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-history.jsonl
The running loss is: 15.809576485306025 The number of items in train is: 26 The loss for epoch 8 0.6080606340502317 3 Stopping model now Data saved to: 28_May_202006_04PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-summary.json
Data saved to:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-history.jsonl
28_May_202006_04PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 529.040771
66 2020-04-22 sub_region ... 66 528.143555
67 2020-04-23 sub_region ... 67 528.838806
68 2020-04-24 sub_region ... 68 528.961365
69 2020-04-25 sub_region ... 69 528.801025
70 2020-04-26 sub_region ... 70 528.515015
71 2020-04-27 sub_region ... 71 528.581177
72 2020-04-28 sub_region ... 72 528.077698
73 2020-04-29 sub_region ... 73 528.772461
74 2020-04-30 sub_region ... 74 528.409302
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/media/plotly/test_plot_18_712b0439.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: bfg54kzp
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/media/plotly/test_plot_all_19_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180408-bfg54kzp/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: bfg54kzp INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 8hcb1kza with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 4 use_mask: True wandb: Agent Started Run: 8hcb1kza
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/8hcb1kza INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjhoY2Ixa3phOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/media/graph/graph_0_summary_963f46d9.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/media/graph
The running loss is: 24.921816915273666 The number of items in train is: 26 The loss for epoch 0 0.958531419818218
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-summary.json
The running loss is: 22.33100463449955 The number of items in train is: 26 The loss for epoch 1 0.858884793634598
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-summary.json
The running loss is: 20.428308714181185 The number of items in train is: 26 The loss for epoch 2 0.7857041813146609
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-summary.json
The running loss is: 17.427796185016632 The number of items in train is: 26 The loss for epoch 3 0.6702998532698705 1
INFO:wandb.wandb_agent:Running runs: ['8hcb1kza']
The running loss is: 17.2829820625484 The number of items in train is: 26 The loss for epoch 4 0.6647300793287846
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-summary.json
2 The running loss is: 15.887216325849295 The number of items in train is: 26 The loss for epoch 5 0.6110467817634344 3 Stopping model now Data saved to: 28_May_202006_04PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-summary.json
Data saved to: 28_May_202006_04PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/config.yaml
Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 559.779724
66 2020-04-22 sub_region ... 66 560.009094
67 2020-04-23 sub_region ... 67 560.145813
68 2020-04-24 sub_region ... 68 560.165466
69 2020-04-25 sub_region ... 69 560.001709
70 2020-04-26 sub_region ... 70 560.061584
71 2020-04-27 sub_region ... 71 559.992859
72 2020-04-28 sub_region ... 72 560.228088
73 2020-04-29 sub_region ... 73 559.984314
74 2020-04-30 sub_region ... 74 560.468567
[21 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/media/plotly/test_plot_12_f72c3f2f.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 8hcb1kza
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/media/plotly/test_plot_all_13_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180424-8hcb1kza/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 8hcb1kza INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 4c6avj80 with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 4 use_mask: False wandb: Agent Started Run: 4c6avj80
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/4c6avj80 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjRjNmF2ajgwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/media/graph/graph_0_summary_4b795b53.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/media
The running loss is: 25.268363758921623 The number of items in train is: 26 The loss for epoch 0 0.9718601445739086
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-history.jsonl
The running loss is: 22.410943418741226 The number of items in train is: 26 The loss for epoch 1 0.8619593622592779
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-history.jsonl
The running loss is: 20.349964793771505 The number of items in train is: 26 The loss for epoch 2 0.7826909536065964
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-history.jsonl
The running loss is: 17.21460761502385 The number of items in train is: 26 The loss for epoch 3 0.6621002928855327 1
INFO:wandb.wandb_agent:Running runs: ['4c6avj80'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-history.jsonl
The running loss is: 17.72370159626007 The number of items in train is: 26 The loss for epoch 4 0.6816808306253873 2 The running loss is: 15.540141738951206 The number of items in train is: 26 The loss for epoch 5 0.597697759190431
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-history.jsonl
3 Stopping model now Data saved to: 28_May_202006_04PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_04PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 559.067444
66 2020-04-22 sub_region ... 66 559.143799
67 2020-04-23 sub_region ... 67 559.442505
68 2020-04-24 sub_region ... 68 559.442566
69 2020-04-25 sub_region ... 69 559.225464
70 2020-04-26 sub_region ... 70 559.264404
71 2020-04-27 sub_region ... 71 559.198975
72 2020-04-28 sub_region ... 72 559.425171
73 2020-04-29 sub_region ... 73 559.358643
74 2020-04-30 sub_region ... 74 559.806274
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/media/plotly/test_plot_12_b8d9f787.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 4c6avj80
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/media/plotly/test_plot_all_13_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180441-4c6avj80/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 4c6avj80 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: dl021jvm with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 5 use_mask: True wandb: Agent Started Run: dl021jvm
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/dl021jvm INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmRsMDIxanZtOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/media/graph/graph_0_summary_a207cf1d.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/media
The running loss is: 20.4773468375206 The number of items in train is: 25 The loss for epoch 0 0.819093873500824
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-summary.json
The running loss is: 19.37137946486473 The number of items in train is: 25 The loss for epoch 1 0.7748551785945892
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-summary.json
The running loss is: 16.317528434097767 The number of items in train is: 25 The loss for epoch 2 0.6527011373639107
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-summary.json
The running loss is: 14.260967433452606 The number of items in train is: 25 The loss for epoch 3 0.5704386973381043 1
INFO:wandb.wandb_agent:Running runs: ['dl021jvm']
The running loss is: 14.17140232771635 The number of items in train is: 25 The loss for epoch 4 0.566856093108654
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-summary.json
The running loss is: 14.054600022733212 The number of items in train is: 25 The loss for epoch 5 0.5621840009093284
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-summary.json
The running loss is: 14.07607826590538 The number of items in train is: 25 The loss for epoch 6 0.5630431306362152
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-summary.json
The running loss is: 14.025826334953308 The number of items in train is: 25 The loss for epoch 7 0.5610330533981324 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-summary.json
The running loss is: 13.500305883586407 The number of items in train is: 25 The loss for epoch 8 0.5400122353434562 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-summary.json
The running loss is: 12.7346723228693 The number of items in train is: 25 The loss for epoch 9 0.509386892914772 3 Stopping model now Data saved to: 28_May_202006_05PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_05PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 384.512299
66 2020-04-22 sub_region ... 66 383.916504
67 2020-04-23 sub_region ... 67 384.334045
68 2020-04-24 sub_region ... 68 384.549500
69 2020-04-25 sub_region ... 69 384.431915
70 2020-04-26 sub_region ... 70 384.146332
71 2020-04-27 sub_region ... 71 384.198578
72 2020-04-28 sub_region ... 72 383.738586
73 2020-04-29 sub_region ... 73 384.381348
74 2020-04-30 sub_region ... 74 383.990662
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/media/plotly/test_plot_20_1987c841.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: dl021jvm
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180458-dl021jvm/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: dl021jvm INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: vk5kcqsx with config: batch_size: 2 forecast_history: 11 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 5 use_mask: False wandb: Agent Started Run: vk5kcqsx
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/vk5kcqsx INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnZrNWtjcXN4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/media/graph/graph_0_summary_f2a463c0.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/media/graph
The running loss is: 20.519908547401428 The number of items in train is: 25 The loss for epoch 0 0.8207963418960571
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-history.jsonl
The running loss is: 19.348501980304718 The number of items in train is: 25 The loss for epoch 1 0.7739400792121888
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-history.jsonl
The running loss is: 16.020245015621185 The number of items in train is: 25 The loss for epoch 2 0.6408098006248474 The running loss is: 14.293965496122837 The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-summary.json
25 The loss for epoch 3
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-history.jsonl
0.5717586198449135 1
INFO:wandb.wandb_agent:Running runs: ['vk5kcqsx']
The running loss is: 14.129456043243408 The number of items in train is: 25 The loss for epoch 4 0.5651782417297363 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-history.jsonl
The running loss is: 13.794427633285522 The number of items in train is: 25 The loss for epoch 5 0.5517771053314209
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-history.jsonl
The running loss is: 13.02604092657566 The number of items in train is: 25 The loss for epoch 6 0.5210416370630264 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-history.jsonl
The running loss is: 9.591085441410542 The number of items in train is: 25 The loss for epoch 7 0.3836434176564217 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-history.jsonl
The running loss is: 7.496750168502331 The number of items in train is: 25 The loss for epoch 8 0.29987000674009323 3 Stopping model now Data saved to: 28_May_202006_05PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_05PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 388.992981
66 2020-04-22 sub_region ... 66 387.531372
67 2020-04-23 sub_region ... 67 388.493744
68 2020-04-24 sub_region ... 68 388.772583
69 2020-04-25 sub_region ... 69 388.547852
70 2020-04-26 sub_region ... 70 387.934692
71 2020-04-27 sub_region ... 71 388.074402
72 2020-04-28 sub_region ... 72 387.295563
73 2020-04-29 sub_region ... 73 388.444427
74 2020-04-30 sub_region ... 74 387.736328
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/media/plotly/test_plot_18_166901d4.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: vk5kcqsx
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/media/plotly/test_plot_all_19_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180519-vk5kcqsx/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: vk5kcqsx INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: k7a8hifp with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 1 use_mask: True wandb: Agent Started Run: k7a8hifp
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/k7a8hifp INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOms3YThoaWZwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
11.178941723890603
The number of items in train is:
27
The loss for epoch 0
0.4140348786626149
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/media/graph/graph_0_summary_c5a40404.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/media
The running loss is: 14.482956442050636 The number of items in train is: 27 The loss for epoch 1 0.5364057941500235 1 The running loss is: 12.16446726070717 The number of items in train is: 27 The loss for epoch 2 0.4505358244706359
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-history.jsonl
The running loss is: 7.8048704912635 The number of items in train is: 27 The loss for epoch 3 0.2890692774542037 1 The running loss is: 7.099390174262226 The number of items in train is: 27 The loss for epoch 4 0.26294037682452687 The running loss is: 6.51896614592988 The number of items in train is: 27 The loss for epoch 5 0.24144319058999558
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-history.jsonl
The running loss is: 6.50308578943077 The number of items in train is: 27 The loss for epoch 6 0.24085502923817667 1 The running loss is: 6.709017640678212 The number of items in train is: 27 The loss for epoch 7 0.24848213483993378 The running loss is: 6.639082251727814 The number of items in train is: 27 The loss for epoch 8 0.2458919352491783 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-history.jsonl
The running loss is: 5.743373372781207 The number of items in train is: 27 The loss for epoch 9 0.21271753232522989 Data saved to: 28_May_202006_05PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_05PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['k7a8hifp'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 758.606506
66 2020-04-22 sub_region ... 66 838.690002
67 2020-04-23 sub_region ... 67 739.158081
68 2020-04-24 sub_region ... 68 760.182251
69 2020-04-25 sub_region ... 69 791.689392
70 2020-04-26 sub_region ... 70 821.668945
71 2020-04-27 sub_region ... 71 813.191467
72 2020-04-28 sub_region ... 72 824.506714
73 2020-04-29 sub_region ... 73 810.240173
74 2020-04-30 sub_region ... 74 760.039307
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: k7a8hifp
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/media/plotly/test_plot_20_50b5ba08.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180538-k7a8hifp/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: k7a8hifp INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 1drnq2sj with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 1 use_mask: False wandb: Agent Started Run: 1drnq2sj
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/1drnq2sj INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjFkcm5xMnNqOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
11.243638909421861
The number of items in train is:
27
The loss for epoch 0
0.41643107071932817
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/media/graph/graph_0_summary_62c56d93.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/media/graph
The running loss is: 15.065432488219813 The number of items in train is: 27 The loss for epoch 1 0.5579789810451783 1 The running loss is: 9.927668149583042 The number of items in train is: 27 The loss for epoch 2 0.36769141294752006
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-history.jsonl
The running loss is: 7.617937269620597 The number of items in train is: 27 The loss for epoch 3 0.2821458248007629 1 The running loss is: 6.723564739339054 The number of items in train is: 27 The loss for epoch 4 0.2490209162718168 The running loss is: 7.1353028768789954 The number of items in train is: 27 The loss for epoch 5 0.2642704769214443 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-history.jsonl
The running loss is: 5.726659771054983 The number of items in train is: 27 The loss for epoch 6 0.21209851003907346 2 The running loss is: 6.842962785856798 The number of items in train is: 27 The loss for epoch 7 0.2534430661428444 The running loss is: 6.147264284198172 The number of items in train is: 27 The loss for epoch 8 0.22767645497030267 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-history.jsonl
The running loss is: 6.772955868189456 The number of items in train is: 27 The loss for epoch 9 0.25085021734035023 Data saved to: 28_May_202006_05PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_05PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['1drnq2sj'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 702.101074
66 2020-04-22 sub_region ... 66 799.573730
67 2020-04-23 sub_region ... 67 688.857788
68 2020-04-24 sub_region ... 68 695.529663
69 2020-04-25 sub_region ... 69 724.923462
70 2020-04-26 sub_region ... 70 761.772095
71 2020-04-27 sub_region ... 71 748.695679
72 2020-04-28 sub_region ... 72 803.100220
73 2020-04-29 sub_region ... 73 733.273804
74 2020-04-30 sub_region ... 74 718.350464
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/media/plotly/test_plot_20_e049cd21.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 1drnq2sj
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180555-1drnq2sj/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 1drnq2sj INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: rrenm2co with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 2 use_mask: True wandb: Agent Started Run: rrenm2co
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/rrenm2co INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnJyZW5tMmNvOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
13.754779417067766
The number of items in train is:
27
The loss for epoch 0
0.5094362747062136
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/media/graph/graph_0_summary_73f15b66.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/media
The running loss is: 15.335549898445606 The number of items in train is: 27 The loss for epoch 1 0.5679833295720594 1 The running loss is: 12.792411483824253 The number of items in train is: 27 The loss for epoch 2 0.4737930179194168
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-summary.json
The running loss is: 10.463905839249492 The number of items in train is: 27 The loss for epoch 3 0.38755206812035153 1 The running loss is: 8.178016487509012 The number of items in train is: 27 The loss for epoch 4 0.3028894995373708 The running loss is: 7.196833549998701 The number of items in train is: 27 The loss for epoch 5 0.2665493907406926 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-summary.json
The running loss is: 7.724783644080162 The number of items in train is: 27 The loss for epoch 6 0.2861030979288949 2 The running loss is: 7.328360717743635 The number of items in train is: 27 The loss for epoch 7 0.27142076732383835 The running loss is: 8.112628399394453 The number of items in train is: 27 The loss for epoch 8 0.30046771849609083 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-summary.json
The running loss is: 6.612887730821967 The number of items in train is: 27 The loss for epoch 9 0.244921767808221 Data saved to: 28_May_202006_06PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_06PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['rrenm2co']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 785.347290
66 2020-04-22 sub_region ... 66 848.155396
67 2020-04-23 sub_region ... 67 778.021973
68 2020-04-24 sub_region ... 68 802.694946
69 2020-04-25 sub_region ... 69 814.782837
70 2020-04-26 sub_region ... 70 827.907227
71 2020-04-27 sub_region ... 71 817.994751
72 2020-04-28 sub_region ... 72 834.097107
73 2020-04-29 sub_region ... 73 812.438049
74 2020-04-30 sub_region ... 74 797.486328
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/media/plotly/test_plot_20_e9ff783e.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: rrenm2co
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180611-rrenm2co/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: rrenm2co INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 0ivbk110 with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 2 use_mask: False wandb: Agent Started Run: 0ivbk110
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/0ivbk110 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjBpdmJrMTEwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
13.334896691143513
The number of items in train is:
27
The loss for epoch 0
0.4938850626349449
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/media/graph/graph_0_summary_4900a83a.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/media/graph
The running loss is: 15.006994970142841 The number of items in train is: 27 The loss for epoch 1 0.5558146285238089 1 The running loss is: 11.964153863489628 The number of items in train is: 27 The loss for epoch 2 0.44311680975887513
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-history.jsonl
The running loss is: 9.46366179548204 The number of items in train is: 27 The loss for epoch 3 0.3505059924252607 1 The running loss is: 7.935978587716818 The number of items in train is: 27 The loss for epoch 4 0.29392513287840066 The running loss is: 7.160205790773034 The number of items in train is: 27 The loss for epoch 5 0.26519280706566795 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-history.jsonl
The running loss is: 7.432532018050551 The number of items in train is: 27 The loss for epoch 6 0.2752789636315019 2 The running loss is: 6.930170862004161 The number of items in train is: 27 The loss for epoch 7 0.256672994889043 The running loss is: 7.583688434679061 The number of items in train is: 27 The loss for epoch 8 0.2808773494325578 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-history.jsonl
The running loss is: 6.852206656709313 The number of items in train is: 27 The loss for epoch 9 0.2537854317299746 Data saved to: 28_May_202006_06PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_06PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['0ivbk110'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 792.838745
66 2020-04-22 sub_region ... 66 839.469238
67 2020-04-23 sub_region ... 67 793.361755
68 2020-04-24 sub_region ... 68 806.429504
69 2020-04-25 sub_region ... 69 816.609619
70 2020-04-26 sub_region ... 70 828.707764
71 2020-04-27 sub_region ... 71 823.823608
72 2020-04-28 sub_region ... 72 829.279419
73 2020-04-29 sub_region ... 73 820.153503
74 2020-04-30 sub_region ... 74 806.989746
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/media/plotly/test_plot_20_4bc95361.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 0ivbk110
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-events.jsonl INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/media/plotly INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180626-0ivbk110/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 0ivbk110 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: djyhq8by with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 3 use_mask: True wandb: Agent Started Run: djyhq8by
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/djyhq8by INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmRqeWhxOGJ5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
18.989742126315832
The number of items in train is:
26
The loss for epoch 0
0.7303746971659936
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/media/graph/graph_0_summary_67e253c4.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/media/graph
The running loss is: 16.38419210910797 The number of items in train is: 26 The loss for epoch 1 0.6301612349656912 1 The running loss is: 15.279926925897598 The number of items in train is: 26 The loss for epoch 2 0.5876894971499076 The running loss is: 10.945733822882175 The number of items in train is: 26 The loss for epoch 3 0.4209897624185452
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-history.jsonl
1 The running loss is: 5.606289468705654 The number of items in train is: 26 The loss for epoch 4 0.21562651802714056 2 The running loss is: 4.256948810070753 The number of items in train is: 26 The loss for epoch 5 0.16372880038733667
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-history.jsonl
The running loss is: 4.631639575585723 The number of items in train is: 26 The loss for epoch 6 0.17813998367637396 1 The running loss is: 6.9152334509417415 The number of items in train is: 26 The loss for epoch 7 0.26597051734391314 The running loss is: 3.967709585092962 The number of items in train is: 26 The loss for epoch 8 0.15260421481126776
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-history.jsonl
The running loss is: 3.5715523255057633 The number of items in train is: 26 The loss for epoch 9 0.13736739713483706 Data saved to: 28_May_202006_06PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_06PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['djyhq8by'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 814.696899
66 2020-04-22 sub_region ... 66 825.249207
67 2020-04-23 sub_region ... 67 843.944824
68 2020-04-24 sub_region ... 68 816.804077
69 2020-04-25 sub_region ... 69 796.358887
70 2020-04-26 sub_region ... 70 799.779358
71 2020-04-27 sub_region ... 71 801.697021
72 2020-04-28 sub_region ... 72 800.930298
73 2020-04-29 sub_region ... 73 815.445740
74 2020-04-30 sub_region ... 74 822.305115
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/media/plotly/test_plot_20_d82f34c9.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: djyhq8by
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180638-djyhq8by/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: djyhq8by INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 0a8lnzah with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 3 use_mask: False wandb: Agent Started Run: 0a8lnzah
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/0a8lnzah INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjBhOGxuemFoOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
18.905425552278757
The number of items in train is:
26
The loss for epoch 0
0.7271317520107214
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/media/graph/graph_0_summary_7ac0ae2a.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/media
The running loss is: 16.324342384934425 The number of items in train is: 26 The loss for epoch 1 0.6278593224974779 1 The running loss is: 15.281524695456028 The number of items in train is: 26 The loss for epoch 2 0.5877509498252318
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-history.jsonl
The running loss is: 10.686566263437271 The number of items in train is: 26 The loss for epoch 3 0.41102177936297196 1 The running loss is: 6.059355471283197 The number of items in train is: 26 The loss for epoch 4 0.2330521335108922 The running loss is: 4.697019984945655 The number of items in train is: 26 The loss for epoch 5 0.1806546148056021
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-history.jsonl
The running loss is: 6.977002476342022 The number of items in train is: 26 The loss for epoch 6 0.26834624909007776 1 The running loss is: 5.771942357998341 The number of items in train is: 26 The loss for epoch 7 0.2219977829999362 The running loss is: 4.5645317947492 The number of items in train is: 26 The loss for epoch 8 0.17555891518266156
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-history.jsonl
The running loss is: 5.686972144991159 The number of items in train is: 26 The loss for epoch 9 0.21872969788427538 1 Data saved to: 28_May_202006_06PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_06PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['0a8lnzah'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 736.851746
66 2020-04-22 sub_region ... 66 772.893066
67 2020-04-23 sub_region ... 67 762.027588
68 2020-04-24 sub_region ... 68 752.332520
69 2020-04-25 sub_region ... 69 736.720947
70 2020-04-26 sub_region ... 70 741.464417
71 2020-04-27 sub_region ... 71 733.462769
72 2020-04-28 sub_region ... 72 760.639526
73 2020-04-29 sub_region ... 73 746.799805
74 2020-04-30 sub_region ... 74 770.807861
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/media/plotly/test_plot_20_556d85cf.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 0a8lnzah
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180655-0a8lnzah/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 0a8lnzah INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 86pghp1t with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 4 use_mask: True wandb: Agent Started Run: 86pghp1t
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/86pghp1t INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjg2cGdocDF0OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
16.948153391480446
The number of items in train is:
26
The loss for epoch 0
0.6518520535184786
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/media/graph/graph_0_summary_30e7d961.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/media/graph
The running loss is: 13.85379159078002 The number of items in train is: 26 The loss for epoch 1 0.5328381381069238 The running loss is: 9.64835274219513 The number of items in train is: 26 The loss for epoch 2 0.37109049008442807 1 The running loss is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-history.jsonl
8.4732926171273
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-summary.json
The number of items in train is: 26 The loss for epoch 3 0.3258958698895115 The running loss is: 5.7244476936757565 The number of items in train is: 26 The loss for epoch 4 0.22017106514137524 The running loss is: 6.5513798370957375 The number of items in train is: 26 The loss for epoch 5 0.25197614758060527 1 The running loss is: 8.37942417897284 The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-history.jsonl
26
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-summary.json
The loss for epoch 6 0.3222855453451092 The running loss is: 4.463485406711698 The number of items in train is: 26 The loss for epoch 7 0.1716725156427576 1 The running loss is: 4.722438246943057 The number of items in train is: 26 The loss for epoch 8 0.18163224026704064
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-summary.json
The running loss is: 4.340588949620724 The number of items in train is: 26 The loss for epoch 9 0.1669457288315663 1 Data saved to: 28_May_202006_07PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_07PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.wandb_agent:Running runs: ['86pghp1t']
CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 851.006592
66 2020-04-22 sub_region ... 66 861.678711
67 2020-04-23 sub_region ... 67 870.574524
68 2020-04-24 sub_region ... 68 854.424133
69 2020-04-25 sub_region ... 69 835.877502
70 2020-04-26 sub_region ... 70 840.583374
71 2020-04-27 sub_region ... 71 839.167969
72 2020-04-28 sub_region ... 72 850.335999
73 2020-04-29 sub_region ... 73 848.938354
74 2020-04-30 sub_region ... 74 859.935547
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/media/plotly/test_plot_20_0f4d1d99.plotly.json
wandb: Agent Finished Run: 86pghp1t
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/media/plotly INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180711-86pghp1t/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 86pghp1t INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: xl3jrbac with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 4 use_mask: False wandb: Agent Started Run: xl3jrbac
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/xl3jrbac INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnhsM2pyYmFjOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
16.634709760546684
The number of items in train is:
26
The loss for epoch 0
0.6397965292517955
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/media/graph/graph_0_summary_cf6f021e.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/media
The running loss is: 14.594469293951988 The number of items in train is: 26 The loss for epoch 1 0.5613257420750765 The running loss is: 9.017393708229065 The number of items in train is: 26 The loss for epoch 2 0.3468228349318871 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-history.jsonl
The running loss is: 7.2338755913078785 The number of items in train is: 26 The loss for epoch 3 0.27822598428107226 The running loss is: 4.978128891438246 The number of items in train is: 26 The loss for epoch 4 0.19146649582454792 1 The running loss is: 5.498803533613682 The number of items in train is: 26 The loss for epoch 5 0.21149244360052621 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-history.jsonl
The running loss is: 7.14033218100667 The number of items in train is: 26 The loss for epoch 6 0.27462816080794883 3 Stopping model now Data saved to: 28_May_202006_07PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_07PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 736.526611
66 2020-04-22 sub_region ... 66 758.064575
67 2020-04-23 sub_region ... 67 751.773193
68 2020-04-24 sub_region ... 68 746.766174
69 2020-04-25 sub_region ... 69 738.920166
70 2020-04-26 sub_region ... 70 743.997314
71 2020-04-27 sub_region ... 71 741.321045
72 2020-04-28 sub_region ... 72 750.091064
73 2020-04-29 sub_region ... 73 743.901367
74 2020-04-30 sub_region ... 74 754.618958
[21 rows x 28 columns]
INFO:wandb.wandb_agent:Running runs: ['xl3jrbac']
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/media/plotly/test_plot_14_251cab71.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: xl3jrbac
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/media/plotly/test_plot_all_15_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180725-xl3jrbac/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: xl3jrbac INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 19jeisii with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 5 use_mask: True wandb: Agent Started Run: 19jeisii
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/19jeisii INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjE5amVpc2lpOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
14.528097435832024
The number of items in train is:
25
The loss for epoch 0
0.581123897433281
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-events.jsonl
The running loss is: 14.321749061346054 The number of items in train is: 25 The loss for epoch 1
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/media/graph/graph_0_summary_d94cc4a3.graph.json
0.5728699624538421
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/media/graph
The running loss is: 8.53690830618143 The number of items in train is: 25 The loss for epoch 2 0.3414763322472572 1 The running loss is: 3.877267986536026 The number of items in train is: 25 The loss for epoch 3 0.15509071946144104
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-history.jsonl
The running loss is: 3.9804319692775607 The number of items in train is: 25 The loss for epoch 4 0.15921727877110242 1 The running loss is: 4.088011069223285 The number of items in train is: 25 The loss for epoch 5 0.16352044276893138 The running loss is: 3.642200954724103 The number of items in train is: 25 The loss for epoch 6 0.14568803818896414 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-history.jsonl
The running loss is: 4.692892727442086 The number of items in train is: 25 The loss for epoch 7 0.18771570909768343 2 The running loss is: 3.2474756240844727 The number of items in train is: 25 The loss for epoch 8 0.1298990249633789 The running loss is: 3.19531544810161 The number of items in train is: 25 The loss for epoch 9 0.1278126179240644 Data saved to: 28_May_202006_07PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-history.jsonl
Data saved to: 28_May_202006_07PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['19jeisii']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 711.413086
66 2020-04-22 sub_region ... 66 714.489502
67 2020-04-23 sub_region ... 67 749.420166
68 2020-04-24 sub_region ... 68 730.427063
69 2020-04-25 sub_region ... 69 707.012146
70 2020-04-26 sub_region ... 70 700.846252
71 2020-04-27 sub_region ... 71 701.634644
72 2020-04-28 sub_region ... 72 702.439697
73 2020-04-29 sub_region ... 73 710.740356
74 2020-04-30 sub_region ... 74 736.382812
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/media/plotly/test_plot_20_b4b65440.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 19jeisii
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180737-19jeisii/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 19jeisii INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: mdyyzx0m with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 1 out_seq_length: 5 use_mask: False wandb: Agent Started Run: mdyyzx0m
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/mdyyzx0m INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm1keXl6eDBtOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
14.583468616008759
The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/config.yaml
25 The loss for epoch 0 0.5833387446403503
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/media/graph/graph_0_summary_67273979.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/media/graph
The running loss is: 11.234198205173016 The number of items in train is: 25 The loss for epoch 1 0.4493679282069206 1 The running loss is: 6.8342317044734955 The number of items in train is: 25 The loss for epoch 2 0.2733692681789398 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-history.jsonl
The running loss is: 4.914771635085344 The number of items in train is: 25 The loss for epoch 3 0.19659086540341378 The running loss is: 6.097113943658769 The number of items in train is: 25 The loss for epoch 4 0.24388455774635076 1 The running loss is: 5.855831284075975 The number of items in train is: 25 The loss for epoch 5 0.23423325136303902 2 The running loss is: 4.823610462248325 The number of items in train is: 25 The loss for epoch 6 0.19294441848993302
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-history.jsonl
The running loss is: 5.116255223751068 The number of items in train is: 25 The loss for epoch 7 0.2046502089500427 1 The running loss is: 3.9354080138728023 The number of items in train is: 25 The loss for epoch 8 0.1574163205549121 2 The running loss is: 3.635712749324739 The number of items in train is: 25 The loss for epoch 9 0.14542850997298956
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-history.jsonl
3 Stopping model now Data saved to: 28_May_202006_07PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_07PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['mdyyzx0m']
interpolate should be below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/config.yaml
Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-history.jsonl
CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 742.793396
66 2020-04-22 sub_region ... 66 736.652344
67 2020-04-23 sub_region ... 67 754.830688
68 2020-04-24 sub_region ... 68 749.173950
69 2020-04-25 sub_region ... 69 738.829346
70 2020-04-26 sub_region ... 70 734.617004
71 2020-04-27 sub_region ... 71 732.794556
72 2020-04-28 sub_region ... 72 736.974976
73 2020-04-29 sub_region ... 73 737.531616
74 2020-04-30 sub_region ... 74 753.737549
[21 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/media/plotly/test_plot_20_71421788.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: mdyyzx0m
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180751-mdyyzx0m/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: mdyyzx0m INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: qyfcznp8 with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 1 use_mask: True wandb: Agent Started Run: qyfcznp8
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/qyfcznp8 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnF5ZmN6bnA4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/media/graph/graph_0_summary_7c646643.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/media/graph
The running loss is: 12.349453534625354 The number of items in train is: 27 The loss for epoch 0 0.4573871679490872 The running loss is: 30.535183822968975 The number of items in train is: 27 The loss for epoch 1 1.130932734184036
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-history.jsonl
The running loss is: 18.304419979453087 The number of items in train is: 27 The loss for epoch 2 0.6779414807204847 1 The running loss is: 10.870923589682207 The number of items in train is: 27
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-summary.json
The loss for epoch 3
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-history.jsonl
0.4026267996178595 2 The running loss is: 9.898105231113732 The number of items in train is: 27 The loss for epoch 4 0.36659649004124933 3 Stopping model now Data saved to: 28_May_202006_08PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_08PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 502.529541
66 2020-04-22 sub_region ... 66 492.594849
67 2020-04-23 sub_region ... 67 496.524628
68 2020-04-24 sub_region ... 68 495.569824
69 2020-04-25 sub_region ... 69 495.434570
70 2020-04-26 sub_region ... 70 493.173859
71 2020-04-27 sub_region ... 71 493.890808
72 2020-04-28 sub_region ... 72 491.183136
73 2020-04-29 sub_region ... 73 497.434570
74 2020-04-30 sub_region ... 74 489.901245
[21 rows x 28 columns]
INFO:wandb.wandb_agent:Running runs: ['qyfcznp8']
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/media/plotly/test_plot_10_c557f1bd.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: qyfcznp8
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/media/plotly/test_plot_all_11_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180802-qyfcznp8/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: qyfcznp8 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: d6rezp1v with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 1 use_mask: False wandb: Agent Started Run: d6rezp1v
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/d6rezp1v INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmQ2cmV6cDF2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/media/graph/graph_0_summary_902a2889.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/media
The running loss is: 12.00125639885664 The number of items in train is: 27 The loss for epoch 0 0.4444909777354311 The running loss is: 27.564760353183374 The number of items in train is: 27 The loss for epoch 1 1.0209170501179028
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-history.jsonl
The running loss is: 19.1493139838567 The number of items in train is: 27 The loss for epoch 2 0.7092338512539519 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-history.jsonl
The running loss is: 9.482985021779314 The number of items in train is: 27 The loss for epoch 3 0.3512216674733079 The running loss is: 8.379846808500588 The number of items in train is: 27 The loss for epoch 4 0.31036469661113286 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-history.jsonl
The running loss is: 6.564052293615532 The number of items in train is: 27 The loss for epoch 5 0.24311304791168636
INFO:wandb.wandb_agent:Running runs: ['d6rezp1v']
The running loss is: 7.043116731874761 The number of items in train is: 27 The loss for epoch 6 0.2608561752546208
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-history.jsonl
The running loss is: 6.5104550529504195 The number of items in train is: 27 The loss for epoch 7 0.2411279649240896 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-history.jsonl
The running loss is: 7.045392430853099 The number of items in train is: 27 The loss for epoch 8 0.2609404604019666 The running loss is: 6.926817309111357 The number of items in train is: 27 The loss for epoch 9 0.25654878922634655 1 Data saved to: 28_May_202006_08PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-history.jsonl
Data saved to: 28_May_202006_08PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 778.043945
66 2020-04-22 sub_region ... 66 783.455933
67 2020-04-23 sub_region ... 67 766.487671
68 2020-04-24 sub_region ... 68 773.625916
69 2020-04-25 sub_region ... 69 782.676147
70 2020-04-26 sub_region ... 70 784.909546
71 2020-04-27 sub_region ... 71 784.748901
72 2020-04-28 sub_region ... 72 778.614258
73 2020-04-29 sub_region ... 73 782.117676
74 2020-04-30 sub_region ... 74 761.960938
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/media/plotly/test_plot_20_647b6c30.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: d6rezp1v
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180814-d6rezp1v/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: d6rezp1v INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: uidcomdz with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 2 use_mask: True wandb: Agent Started Run: uidcomdz
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/uidcomdz INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnVpZGNvbWR6OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/media/graph/graph_0_summary_9d362d29.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/media/graph
The running loss is: 15.362042212858796 The number of items in train is: 27 The loss for epoch 0 0.5689645264021777 The running loss is: 18.604257106781006 The number of items in train is: 27 The loss for epoch 1 0.6890465595104076 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-history.jsonl
The running loss is: 19.387704892084002 The number of items in train is: 27 The loss for epoch 2 0.7180631441512594 The running loss is: 17.115502156317234 The number of items in train is: 27
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-summary.json
The loss for epoch 3 0.6339074872710087
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-history.jsonl
The running loss is: 9.399747041985393 The number of items in train is: 27 The loss for epoch 4 0.3481387793327923 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-history.jsonl
The running loss is: 8.062234118580818 The number of items in train is: 27 The loss for epoch 5 0.2986012636511414
INFO:wandb.wandb_agent:Running runs: ['uidcomdz']
The running loss is: 7.508644045330584 The number of items in train is: 27 The loss for epoch 6 0.2780979276048364
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-history.jsonl
The running loss is: 7.624856340698898 The number of items in train is: 27 The loss for epoch 7 0.2824020866925518 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-history.jsonl
The running loss is: 7.4580579325556755 The number of items in train is: 27 The loss for epoch 8 0.27622436787243243 The running loss is: 6.958448266610503 The number of items in train is: 27 The loss for epoch 9 0.25772030617075936 Data saved to: 28_May_202006_08PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-history.jsonl
Data saved to: 28_May_202006_08PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 772.555298
66 2020-04-22 sub_region ... 66 766.406433
67 2020-04-23 sub_region ... 67 773.479492
68 2020-04-24 sub_region ... 68 770.044922
69 2020-04-25 sub_region ... 69 772.980347
70 2020-04-26 sub_region ... 70 772.595215
71 2020-04-27 sub_region ... 71 772.935425
72 2020-04-28 sub_region ... 72 770.757202
73 2020-04-29 sub_region ... 73 772.154602
74 2020-04-30 sub_region ... 74 771.334900
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: uidcomdz
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/media/plotly/test_plot_20_cf3ed598.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180831-uidcomdz/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: uidcomdz INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: gezep2gw with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 2 use_mask: False wandb: Agent Started Run: gezep2gw
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/gezep2gw INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmdlemVwMmd3OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/media/graph/graph_0_summary_69bc7e29.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/media
The running loss is: 15.227613762952387 The number of items in train is: 27 The loss for epoch 0 0.5639856949241625 The running loss is: 17.898133425042033 The number of items in train is: 27 The loss for epoch 1 0.6628938305571124 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-summary.json
The running loss is: 15.654380347579718 The number of items in train is: 27 The loss for epoch 2 0.5797918647251747
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-summary.json
The running loss is: 9.89348728954792 The number of items in train is: 27 The loss for epoch 3 0.3664254551684415 1 The running loss is: 8.431944162817672 The number of items in train is: 27 The loss for epoch 4 0.31229422825250636
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-summary.json
The running loss is: 8.609166414942592 The number of items in train is: 27 The loss for epoch 5 0.31885801536824415 1
INFO:wandb.wandb_agent:Running runs: ['gezep2gw']
The running loss is: 9.016866168007255 The number of items in train is: 27 The loss for epoch 6 0.3339580062224909 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-summary.json
The running loss is: 8.0558643322438 The number of items in train is: 27 The loss for epoch 7 0.29836534563865924
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-summary.json
The running loss is: 8.685117572546005 The number of items in train is: 27 The loss for epoch 8 0.3216710212054076 1 The running loss is: 6.745234100148082 The number of items in train is: 27 The loss for epoch 9 0.2498234851906697 Data saved to: 28_May_202006_08PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-summary.json
Data saved to: 28_May_202006_08PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 783.742432
66 2020-04-22 sub_region ... 66 782.695801
67 2020-04-23 sub_region ... 67 783.955322
68 2020-04-24 sub_region ... 68 783.325439
69 2020-04-25 sub_region ... 69 783.187744
70 2020-04-26 sub_region ... 70 783.111938
71 2020-04-27 sub_region ... 71 783.345886
72 2020-04-28 sub_region ... 72 782.962280
73 2020-04-29 sub_region ... 73 782.751831
74 2020-04-30 sub_region ... 74 783.127808
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/media/plotly/test_plot_20_86f33ef2.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: gezep2gw
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180848-gezep2gw/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: gezep2gw INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 3xj34yik with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 3 use_mask: True wandb: Agent Started Run: 3xj34yik
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/3xj34yik INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjN4ajM0eWlrOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/media/graph/graph_0_summary_0da95377.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/media
The running loss is: 21.012039236724377 The number of items in train is: 26 The loss for epoch 0 0.8081553552586299 The running loss is: 16.833910040557384 The number of items in train is: 26 The loss for epoch 1 0.6474580784829763 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-summary.json
The running loss is: 16.23337048664689 The number of items in train is: 26 The loss for epoch 2 0.6243604033325727 The running loss is: 15.538650223985314 The number of items in train is: 26 The loss for epoch 3 0.5976403932302043
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-summary.json
1 The running loss is: 15.044064596295357 The number of items in train is: 26 The loss for epoch 4 0.5786178690882829 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-summary.json
The running loss is: 11.831363804638386 The number of items in train is: 26 The loss for epoch 5 0.4550524540245533 The running loss is: 7.703582746908069 The number of items in train is: 26 The loss for epoch 6 0.2962916441118488
INFO:wandb.wandb_agent:Running runs: ['3xj34yik']
1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-summary.json
The running loss is: 8.853043738752604 The number of items in train is: 26 The loss for epoch 7 0.3405016822597155 2 The running loss is: 5.49346605874598 The number of items in train is: 26 The loss for epoch 8 0.21128715610561463
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-summary.json
The running loss is: 6.049623145721853 The number of items in train is: 26 The loss for epoch 9 0.23267781329699433 Data saved to: 28_May_202006_09PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_09PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 793.348389
66 2020-04-22 sub_region ... 66 793.042908
67 2020-04-23 sub_region ... 67 790.546631
68 2020-04-24 sub_region ... 68 791.214233
69 2020-04-25 sub_region ... 69 792.006836
70 2020-04-26 sub_region ... 70 792.608032
71 2020-04-27 sub_region ... 71 792.963989
72 2020-04-28 sub_region ... 72 789.634521
73 2020-04-29 sub_region ... 73 793.287476
74 2020-04-30 sub_region ... 74 785.882202
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/media/plotly/test_plot_20_ef160644.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 3xj34yik
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180905-3xj34yik/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 3xj34yik INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: m7imp1fv with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 3 use_mask: False wandb: Agent Started Run: m7imp1fv
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/m7imp1fv INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm03aW1wMWZ2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/requirements.txt
The running loss is: 20.714700505137444
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/wandb-events.jsonl
The number of items in train is: 26
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/media/graph/graph_0_summary_1c6cc7d6.graph.json
The loss for epoch 0
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/wandb-summary.json
0.7967192501975939
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/media
The running loss is: 16.891354955732822 The number of items in train is: 26 The loss for epoch 1 0.6496674982974162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/wandb-history.jsonl
The running loss is: 16.139072712510824 The number of items in train is: 26 The loss for epoch 2 0.620733565865801 1 The running loss is: 15.367559241130948
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/wandb-summary.json
The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/wandb-history.jsonl
26 The loss for epoch 3 0.5910599708127288 2 The running loss is: 14.148938469588757 The number of items in train is: 26 The loss for epoch 4 0.5441899411380291 3 Stopping model now Data saved to: 28_May_202006_09PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/wandb-history.jsonl
Data saved to: 28_May_202006_09PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
INFO:wandb.wandb_agent:Running runs: ['m7imp1fv']
date level ... original_index preds 54 2020-04-10 sub_region ... 54 0.000000 55 2020-04-11 sub_region ... 55 0.000000 56 2020-04-12 sub_region ... 56 0.000000 57 2020-04-13 sub_region ... 57 0.000000 58 2020-04-14 sub_region ... 58 0.000000 59 2020-04-15 sub_region ... 59 0.000000 60 2020-04-16 sub_region ... 60 0.000000 61 2020-04-17 sub_region ... 61 0.000000 62 2020-04-18 sub_region ... 62 0.000000 63 2020-04-19 sub_region ... 63 0.000000 64 2020-04-20 sub_region ... 64 0.000000 65 2020-04-21 sub_region ... 65 440.134735 66 2020-04-22 sub_region ... 66 439.800903 67 2020-04-23 sub_region ... 67 443.283325 68 2020-04-24 sub_region ... 68 442.842438 69 2020-04-25 sub_region ... 69 440.653381 70 2020-04-26 sub_region ... 70 441.242767 71 2020-04-27 sub_region ... 71 440.577118 72 2020-04-28 sub_region ... 72 442.682800 73 2020-04-29 sub_region ... 73 441.382904 74 2020-04-30 sub_region ... 74 445.453522 [21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: m7imp1fv
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/media/plotly/test_plot_10_0d49373a.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/media/plotly/test_plot_all_11_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180922-m7imp1fv/wandb-events.jsonl INFO:wandb.wandb_agent:Cleaning up finished run: m7imp1fv INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ecwpsoe9 with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 4 use_mask: True wandb: Agent Started Run: ecwpsoe9
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ecwpsoe9 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmVjd3Bzb2U5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/media/graph/graph_0_summary_54d34cc7.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/media
The running loss is: 23.482627481222153 The number of items in train is: 26 The loss for epoch 0 0.9031779800470059 The running loss is: 20.23513476550579 The number of items in train is: 26 The loss for epoch 1 0.778274414057915 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-summary.json
The running loss is: 21.649361565709114 The number of items in train is: 26 The loss for epoch 2 0.8326677525272737 The running loss is: 15.995433270931244 The number of items in train is: 26 The loss for epoch 3 0.615208971958894
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-summary.json
1 The running loss is: 12.781472019851208 The number of items in train is: 26 The loss for epoch 4 0.4915950776865849 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-summary.json
The running loss is: 8.850842893123627 The number of items in train is: 26 The loss for epoch 5 0.3404170343509087 The running loss is: 6.8902468085289 The number of items in train is: 26 The loss for epoch 6 0.26500949263572693
INFO:wandb.wandb_agent:Running runs: ['ecwpsoe9']
1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-summary.json
The running loss is: 8.579151637852192 The number of items in train is: 26 The loss for epoch 7 0.32996737068662274 2 The running loss is: 6.014725340530276 The number of items in train is: 26 The loss for epoch 8 0.23133559002039525
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-summary.json
The running loss is: 4.920796798542142 The number of items in train is: 26 The loss for epoch 9 0.18926141532854393 1 Data saved to: 28_May_202006_09PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_09PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 784.844604
66 2020-04-22 sub_region ... 66 785.061829
67 2020-04-23 sub_region ... 67 785.951660
68 2020-04-24 sub_region ... 68 785.422729
69 2020-04-25 sub_region ... 69 784.747925
70 2020-04-26 sub_region ... 70 784.533691
71 2020-04-27 sub_region ... 71 784.408447
72 2020-04-28 sub_region ... 72 785.171936
73 2020-04-29 sub_region ... 73 784.649170
74 2020-04-30 sub_region ... 74 785.943665
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/media/plotly/test_plot_20_bf33d510.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ecwpsoe9
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180934-ecwpsoe9/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ecwpsoe9 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 27b1l264 with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 4 use_mask: False wandb: Agent Started Run: 27b1l264
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/27b1l264 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjI3YjFsMjY0OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/media/graph/graph_0_summary_9c42a37e.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/media
The running loss is: 23.31422958523035 The number of items in train is: 26 The loss for epoch 0 0.896701137893475 The running loss is: 19.628748193383217 The number of items in train is: 26 The loss for epoch 1 0.7549518535916622 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-history.jsonl
The running loss is: 22.03840383887291 The number of items in train is: 26 The loss for epoch 2 0.8476309168797272
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-history.jsonl
The running loss is: 15.539160393178463 The number of items in train is: 26 The loss for epoch 3 0.5976600151222485 1 The running loss is: 12.818177118897438 The number of items in train is: 26 The loss for epoch 4 0.4930068122652861
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-history.jsonl
The running loss is: 10.63734956085682 The number of items in train is: 26 The loss for epoch 5 0.4091288292637238 1
INFO:wandb.wandb_agent:Running runs: ['27b1l264'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-history.jsonl
The running loss is: 6.949803344905376 The number of items in train is: 26 The loss for epoch 6 0.2673001286502068
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-history.jsonl
The running loss is: 7.64347056671977 The number of items in train is: 26 The loss for epoch 7 0.29397963718152964 1 The running loss is: 11.507175385951996 The number of items in train is: 26 The loss for epoch 8 0.4425836686904614
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-history.jsonl
The running loss is: 5.667488571256399 The number of items in train is: 26 The loss for epoch 9 0.21798032966370767 Data saved to: 28_May_202006_09PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_09PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 749.537842
66 2020-04-22 sub_region ... 66 753.229736
67 2020-04-23 sub_region ... 67 753.515015
68 2020-04-24 sub_region ... 68 753.205444
69 2020-04-25 sub_region ... 69 751.122925
70 2020-04-26 sub_region ... 70 750.473755
71 2020-04-27 sub_region ... 71 748.162842
72 2020-04-28 sub_region ... 72 752.245361
73 2020-04-29 sub_region ... 73 750.474243
74 2020-04-30 sub_region ... 74 754.320679
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 27b1l264
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/media/plotly/test_plot_20_67e12e5e.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_180951-27b1l264/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 27b1l264 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 8g69sz1i with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 5 use_mask: True wandb: Agent Started Run: 8g69sz1i
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/8g69sz1i INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjhnNjlzejFpOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/media/graph/graph_0_summary_1c2050d9.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/media
The running loss is: 17.830005079507828 The number of items in train is: 25 The loss for epoch 0 0.7132002031803131 The running loss is: 16.9639795422554 The number of items in train is: 25 The loss for epoch 1 0.6785591816902161 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-history.jsonl
The running loss is: 19.417406290769577 The number of items in train is: 25 The loss for epoch 2 0.776696251630783 The running loss is: 13.737709395587444 The number of items in train is: 25 The loss for epoch 3 0.5495083758234978
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-history.jsonl
1 The running loss is: 13.103170596063137 The number of items in train is: 25 The loss for epoch 4 0.5241268238425255 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-history.jsonl
The running loss is: 9.996175795793533 The number of items in train is: 25 The loss for epoch 5 0.3998470318317413 3 Stopping model now Data saved to: 28_May_202006_10PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_10PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['8g69sz1i']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-summary.json
date level ... original_index preds 54 2020-04-10 sub_region ... 54 0.000000 55 2020-04-11 sub_region ... 55 0.000000 56 2020-04-12 sub_region ... 56 0.000000 57 2020-04-13 sub_region ... 57 0.000000 58 2020-04-14 sub_region ... 58 0.000000 59 2020-04-15 sub_region ... 59 0.000000 60 2020-04-16 sub_region ... 60 0.000000 61 2020-04-17 sub_region ... 61 0.000000 62 2020-04-18 sub_region ... 62 0.000000 63 2020-04-19 sub_region ... 63 0.000000 64 2020-04-20 sub_region ... 64 0.000000 65 2020-04-21 sub_region ... 65 417.858063 66 2020-04-22 sub_region ... 66 419.026245 67 2020-04-23 sub_region ... 67 420.017090 68 2020-04-24 sub_region ... 68 420.572571 69 2020-04-25 sub_region ... 69 419.899597 70 2020-04-26 sub_region ... 70 419.641113 71 2020-04-27 sub_region ... 71 419.378143 72 2020-04-28 sub_region ... 72 419.695251 73 2020-04-29 sub_region ... 73 419.175232 74 2020-04-30 sub_region ... 74 421.516815 [21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/media/plotly/test_plot_12_d17e53a2.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 8g69sz1i
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/media/plotly/test_plot_all_13_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181008-8g69sz1i/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 8g69sz1i INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: xxokbmwc with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 2 out_seq_length: 5 use_mask: False wandb: Agent Started Run: xxokbmwc
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/xxokbmwc INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnh4b2tibXdjOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/media/graph/graph_0_summary_6547bba0.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/media
The running loss is:
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/media/graph
17.741086795926094 The number of items in train is: 25 The loss for epoch 0 0.7096434718370438 The running loss is: 16.730679839849472 The number of items in train is: 25 The loss for epoch 1 0.6692271935939789 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-summary.json
The running loss is: 19.939407482743263 The number of items in train is: 25 The loss for epoch 2 0.7975762993097305 The running loss is: 13.255669929087162 The number of items in train is: 25 The loss for epoch 3 0.5302267971634865 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-summary.json
The running loss is: 11.027616888284683 The number of items in train is: 25 The loss for epoch 4 0.4411046755313873 2 The running loss is: 6.718288138508797 The number of items in train is: 25 The loss for epoch 5 0.26873152554035185
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-summary.json
The running loss is: 7.313029149547219 The number of items in train is: 25 The loss for epoch 6 0.29252116598188876 1
INFO:wandb.wandb_agent:Running runs: ['xxokbmwc'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-summary.json
The running loss is: 12.773601979017258 The number of items in train is: 25 The loss for epoch 7 0.5109440791606903 The running loss is: 6.80379575304687 The number of items in train is: 25 The loss for epoch 8 0.2721518301218748 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-summary.json
The running loss is: 4.739170776680112 The number of items in train is: 25 The loss for epoch 9 0.18956683106720448 2 Data saved to: 28_May_202006_10PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_10PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 770.853760
66 2020-04-22 sub_region ... 66 769.976440
67 2020-04-23 sub_region ... 67 771.220154
68 2020-04-24 sub_region ... 68 770.911987
69 2020-04-25 sub_region ... 69 770.517700
70 2020-04-26 sub_region ... 70 770.186768
71 2020-04-27 sub_region ... 71 770.193970
72 2020-04-28 sub_region ... 72 770.157593
73 2020-04-29 sub_region ... 73 770.232788
74 2020-04-30 sub_region ... 74 770.960083
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/media/plotly/test_plot_20_57ec849c.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: xxokbmwc
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181020-xxokbmwc/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: xxokbmwc INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: srirmbfg with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 1 use_mask: True wandb: Agent Started Run: srirmbfg
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/srirmbfg INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnNyaXJtYmZnOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/media/graph/graph_0_summary_236fd991.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/media/graph
The running loss is: 13.64892402663827 The number of items in train is: 27 The loss for epoch 0 0.5055157046903063
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-history.jsonl
The running loss is: 37.692751033231616 The number of items in train is: 27 The loss for epoch 1 1.3960278160456154
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-history.jsonl
The running loss is: 20.47567129507661 The number of items in train is: 27 The loss for epoch 2 0.7583581961139485
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-history.jsonl
The running loss is: 20.933722829446197 The number of items in train is: 27 The loss for epoch 3 0.7753230677572666 1
INFO:wandb.wandb_agent:Running runs: ['srirmbfg'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-history.jsonl
The running loss is: 20.798208009451628 The number of items in train is: 27 The loss for epoch 4 0.7703040003500603 The running loss is: 20.184429235756397 The number of items in train is: 27 The loss for epoch 5 0.7475714531761629
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-history.jsonl
1 The running loss is: 20.411993622779846 The number of items in train is: 27 The loss for epoch 6 0.755999763806661 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-history.jsonl
The running loss is: 19.94804622977972 The number of items in train is: 27 The loss for epoch 7 0.7388165270288786 3 Stopping model now Data saved to: 28_May_202006_10PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-history.jsonl
Data saved to: 28_May_202006_10PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 474.087341
66 2020-04-22 sub_region ... 66 474.107361
67 2020-04-23 sub_region ... 67 473.755005
68 2020-04-24 sub_region ... 68 474.003113
69 2020-04-25 sub_region ... 69 474.199402
70 2020-04-26 sub_region ... 70 474.221405
71 2020-04-27 sub_region ... 71 474.191833
72 2020-04-28 sub_region ... 72 473.996887
73 2020-04-29 sub_region ... 73 474.185181
74 2020-04-30 sub_region ... 74 473.697998
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/media/plotly/test_plot_16_1f8d7214.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: srirmbfg
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/media/plotly/test_plot_all_17_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181037-srirmbfg/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: srirmbfg INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: t4xosvtx with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 1 use_mask: False wandb: Agent Started Run: t4xosvtx
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/t4xosvtx INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnQ0eG9zdnR4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/media/graph/graph_0_summary_453d443c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/media
The running loss is: 13.37478191871196 The number of items in train is: 27 The loss for epoch 0 0.49536229328562814
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-history.jsonl
The running loss is: 37.45775743201375 The number of items in train is: 27 The loss for epoch 1 1.3873243493338425
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-history.jsonl
The running loss is: 20.595209177583456 The number of items in train is: 27 The loss for epoch 2 0.7627855250956835
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-history.jsonl
The running loss is: 21.006036780774593 The number of items in train is: 27 The loss for epoch 3 0.7780013622509109
INFO:wandb.wandb_agent:Running runs: ['t4xosvtx'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-history.jsonl
The running loss is: 20.73020259477198 The number of items in train is: 27 The loss for epoch 4 0.7677852812878512 The running loss is: 20.24301341921091 The number of items in train is: 27 The loss for epoch 5 0.7497412377485523
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-history.jsonl
1 The running loss is: 20.167157787829638 The number of items in train is: 27 The loss for epoch 6 0.7469317699196162 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-history.jsonl
The running loss is: 18.895125970244408 The number of items in train is: 27 The loss for epoch 7 0.6998194803794225 3 Stopping model now Data saved to: 28_May_202006_11PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-history.jsonl
Data saved to: 28_May_202006_11PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 474.395355
66 2020-04-22 sub_region ... 66 474.474243
67 2020-04-23 sub_region ... 67 473.946503
68 2020-04-24 sub_region ... 68 474.251740
69 2020-04-25 sub_region ... 69 474.498688
70 2020-04-26 sub_region ... 70 474.531128
71 2020-04-27 sub_region ... 71 474.499969
72 2020-04-28 sub_region ... 72 474.257050
73 2020-04-29 sub_region ... 73 474.527222
74 2020-04-30 sub_region ... 74 473.818420
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: t4xosvtx
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/media/plotly/test_plot_all_17_6655637c.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/media/plotly/test_plot_16_c257944a.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181058-t4xosvtx/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: t4xosvtx INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: clrrnsx0 with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 2 use_mask: True wandb: Agent Started Run: clrrnsx0
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/clrrnsx0 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmNscnJuc3gwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/media/graph/graph_0_summary_1eb04cf7.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/media
The running loss is: 15.137208757922053 The number of items in train is: 27 The loss for epoch 0 0.5606373614045205
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-history.jsonl
The running loss is: 30.88287841156125 The number of items in train is: 27 The loss for epoch 1 1.1438103115393057
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-history.jsonl
The running loss is: 25.371903955936432 The number of items in train is: 27 The loss for epoch 2 0.9397001465161642
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-history.jsonl
The running loss is: 20.891498629003763 The number of items in train is: 27 The loss for epoch 3 0.7737592084816208
INFO:wandb.wandb_agent:Running runs: ['clrrnsx0']
The running loss is: 20.58536821603775 The number of items in train is: 27 The loss for epoch 4 0.7624210450384352
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-history.jsonl
1 The running loss is: 21.16692252457142 The number of items in train is: 27 The loss for epoch 5 0.7839600935026452 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-history.jsonl
The running loss is: 20.775313809514046 The number of items in train is: 27 The loss for epoch 6 0.7694560670190387 3 Stopping model now Data saved to: 28_May_202006_11PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-history.jsonl
Data saved to: 28_May_202006_11PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 422.063049
66 2020-04-22 sub_region ... 66 422.065277
67 2020-04-23 sub_region ... 67 422.008148
68 2020-04-24 sub_region ... 68 422.044128
69 2020-04-25 sub_region ... 69 422.075989
70 2020-04-26 sub_region ... 70 422.085114
71 2020-04-27 sub_region ... 71 422.084900
72 2020-04-28 sub_region ... 72 422.078888
73 2020-04-29 sub_region ... 73 422.060028
74 2020-04-30 sub_region ... 74 422.037537
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/media/plotly/test_plot_14_7a77c25e.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/media/plotly
wandb: Agent Finished Run: clrrnsx0
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/media/plotly/test_plot_all_15_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181114-clrrnsx0/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: clrrnsx0 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: aiej4x17 with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 2 use_mask: False wandb: Agent Started Run: aiej4x17
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/aiej4x17 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmFpZWo0eDE3OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/media/graph/graph_0_summary_83359f6b.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/media/graph
The running loss is: 15.071377269923687 The number of items in train is: 27 The loss for epoch 0 0.5581991581453217
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-history.jsonl
The running loss is: 31.568736284971237 The number of items in train is: 27 The loss for epoch 1 1.1692124549989347
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-history.jsonl
The running loss is: 25.198651179671288 The number of items in train is: 27 The loss for epoch 2 0.9332833770248625
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-history.jsonl
The running loss is: 21.102718703448772 The number of items in train is: 27 The loss for epoch 3 0.7815821742018064
INFO:wandb.wandb_agent:Running runs: ['aiej4x17']
The running loss is: 20.57193885743618 The number of items in train is: 27 The loss for epoch 4 0.7619236613865252
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-history.jsonl
1 The running loss is: 21.114420972764492 The number of items in train is: 27 The loss for epoch 5 0.7820155915838701 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-history.jsonl
The running loss is: 21.067522056400776 The number of items in train is: 27 The loss for epoch 6 0.7802785946815102 3 Stopping model now Data saved to: 28_May_202006_11PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-history.jsonl
Data saved to: 28_May_202006_11PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 423.483063
66 2020-04-22 sub_region ... 66 423.511169
67 2020-04-23 sub_region ... 67 423.472900
68 2020-04-24 sub_region ... 68 423.498932
69 2020-04-25 sub_region ... 69 423.520294
70 2020-04-26 sub_region ... 70 423.524902
71 2020-04-27 sub_region ... 71 423.520996
72 2020-04-28 sub_region ... 72 423.510651
73 2020-04-29 sub_region ... 73 423.502411
74 2020-04-30 sub_region ... 74 423.487152
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/media/plotly/test_plot_14_0bb7c432.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: aiej4x17
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/media/plotly/test_plot_all_15_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181131-aiej4x17/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: aiej4x17 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: bt7r2ilt with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 3 use_mask: True wandb: Agent Started Run: bt7r2ilt
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/bt7r2ilt INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmJ0N3IyaWx0OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/media/graph/graph_0_summary_e4022ec7.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/media
The running loss is: 24.238217145204544 The number of items in train is: 26 The loss for epoch 0 0.9322391209694055
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-summary.json
The running loss is: 16.11776690930128 The number of items in train is: 26 The loss for epoch 1 0.6199141118962032 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-summary.json
The running loss is: 16.947837613523006 The number of items in train is: 26 The loss for epoch 2 0.6518399082124233 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-summary.json
The running loss is: 17.64614214003086 The number of items in train is: 26 The loss for epoch 3 0.6786977746165715
INFO:wandb.wandb_agent:Running runs: ['bt7r2ilt']
The running loss is: 16.301386021077633 The number of items in train is: 26 The loss for epoch 4 0.6269763854260628
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-summary.json
The running loss is: 16.042810605838895 The number of items in train is: 26 The loss for epoch 5 0.6170311771476498
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-summary.json
The running loss is: 15.744363751262426 The number of items in train is: 26 The loss for epoch 6 0.6055524519716318 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-summary.json
The running loss is: 16.016306111589074 The number of items in train is: 26 The loss for epoch 7 0.6160117735226567 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-summary.json
The running loss is: 15.813427444547415 The number of items in train is: 26 The loss for epoch 8 0.6082087478672082
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-summary.json
The running loss is: 16.24589281156659 The number of items in train is: 26 The loss for epoch 9 0.6248420312140996 Data saved to: 28_May_202006_11PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_11PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 509.934052
66 2020-04-22 sub_region ... 66 507.839417
67 2020-04-23 sub_region ... 67 509.408234
68 2020-04-24 sub_region ... 68 509.694031
69 2020-04-25 sub_region ... 69 509.575317
70 2020-04-26 sub_region ... 70 508.830750
71 2020-04-27 sub_region ... 71 509.065674
72 2020-04-28 sub_region ... 72 507.174622
73 2020-04-29 sub_region ... 73 509.588074
74 2020-04-30 sub_region ... 74 507.856110
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/media/plotly/test_plot_20_75179c3f.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: bt7r2ilt
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181149-bt7r2ilt/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: bt7r2ilt INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: d4g2l0ez with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 3 use_mask: False wandb: Agent Started Run: d4g2l0ez
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/d4g2l0ez INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmQ0ZzJsMGV6OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/media/graph/graph_0_summary_1e552aae.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/media
The running loss is: 24.46081606298685 The number of items in train is: 26 The loss for epoch 0 0.9408006178071866
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-history.jsonl
The running loss is: 16.266936726868153 The number of items in train is: 26 The loss for epoch 1 0.625651412571852 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-history.jsonl
The running loss is: 16.894336327910423 The number of items in train is: 26 The loss for epoch 2 0.6497821664580932
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-history.jsonl
The running loss is: 17.256735399365425 The number of items in train is: 26 The loss for epoch 3 0.6637205922832856
INFO:wandb.wandb_agent:Running runs: ['d4g2l0ez']
The running loss is: 16.231802832335234 The number of items in train is: 26 The loss for epoch 4 0.6243001089359705
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-history.jsonl
The running loss is: 16.1003286074847 The number of items in train is: 26 The loss for epoch 5 0.6192434079801807
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-history.jsonl
The running loss is: 15.810061607509851 The number of items in train is: 26 The loss for epoch 6 0.6080792925965327 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-history.jsonl
The running loss is: 16.048187194392085 The number of items in train is: 26 The loss for epoch 7 0.6172379690150802 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-history.jsonl
The running loss is: 15.809576485306025 The number of items in train is: 26 The loss for epoch 8 0.6080606340502317 3 Stopping model now Data saved to: 28_May_202006_12PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_12PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 529.040771
66 2020-04-22 sub_region ... 66 528.143555
67 2020-04-23 sub_region ... 67 528.838806
68 2020-04-24 sub_region ... 68 528.961365
69 2020-04-25 sub_region ... 69 528.801025
70 2020-04-26 sub_region ... 70 528.515015
71 2020-04-27 sub_region ... 71 528.581177
72 2020-04-28 sub_region ... 72 528.077698
73 2020-04-29 sub_region ... 73 528.772461
74 2020-04-30 sub_region ... 74 528.409302
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/media/plotly/test_plot_18_712b0439.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: d4g2l0ez
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/media/plotly/test_plot_all_19_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181215-d4g2l0ez/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: d4g2l0ez INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 89ypi0wm with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 4 use_mask: True wandb: Agent Started Run: 89ypi0wm
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/89ypi0wm INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjg5eXBpMHdtOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/media/graph/graph_0_summary_7aec0b6e.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/media
The running loss is: 24.921816915273666 The number of items in train is: 26 The loss for epoch 0 0.958531419818218
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-history.jsonl
The running loss is: 22.33100463449955 The number of items in train is: 26 The loss for epoch 1 0.858884793634598
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-history.jsonl
The running loss is: 20.428308714181185 The number of items in train is: 26 The loss for epoch 2 0.7857041813146609
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-history.jsonl
The running loss is: 17.427796185016632 The number of items in train is: 26 The loss for epoch 3 0.6702998532698705 1
INFO:wandb.wandb_agent:Running runs: ['89ypi0wm']
The running loss is: 17.2829820625484 The number of items in train is: 26 The loss for epoch 4 0.6647300793287846
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-history.jsonl
2 The running loss is: 15.887216325849295 The number of items in train is: 26 The loss for epoch 5 0.6110467817634344 3 Stopping model now Data saved to: 28_May_202006_12PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-history.jsonl
Data saved to: 28_May_202006_12PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 559.779724
66 2020-04-22 sub_region ... 66 560.009094
67 2020-04-23 sub_region ... 67 560.145813
68 2020-04-24 sub_region ... 68 560.165466
69 2020-04-25 sub_region ... 69 560.001709
70 2020-04-26 sub_region ... 70 560.061584
71 2020-04-27 sub_region ... 71 559.992859
72 2020-04-28 sub_region ... 72 560.228088
73 2020-04-29 sub_region ... 73 559.984314
74 2020-04-30 sub_region ... 74 560.468567
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/media/plotly/test_plot_12_f72c3f2f.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 89ypi0wm
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/media/plotly/test_plot_all_13_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181237-89ypi0wm/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 89ypi0wm INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: a9hey38z with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 4 use_mask: False wandb: Agent Started Run: a9hey38z
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/a9hey38z INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmE5aGV5Mzh6OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/media/graph/graph_0_summary_8cd356a7.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/media
The running loss is: 25.268363758921623 The number of items in train is: 26 The loss for epoch 0 0.9718601445739086
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-history.jsonl
The running loss is: 22.410943418741226 The number of items in train is: 26 The loss for epoch 1 0.8619593622592779
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-history.jsonl
The running loss is: 20.349964793771505 The number of items in train is: 26 The loss for epoch 2 0.7826909536065964
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-history.jsonl
The running loss is: 17.21460761502385 The number of items in train is: 26 The loss for epoch 3 0.6621002928855327 1
INFO:wandb.wandb_agent:Running runs: ['a9hey38z']
The running loss is: 17.72370159626007 The number of items in train is: 26 The loss for epoch 4 0.6816808306253873 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-history.jsonl
The running loss is: 15.540141738951206 The number of items in train is: 26 The loss for epoch 5 0.597697759190431 3 Stopping model now Data saved to: 28_May_202006_13PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-history.jsonl
Data saved to: 28_May_202006_13PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 559.067444
66 2020-04-22 sub_region ... 66 559.143799
67 2020-04-23 sub_region ... 67 559.442505
68 2020-04-24 sub_region ... 68 559.442566
69 2020-04-25 sub_region ... 69 559.225464
70 2020-04-26 sub_region ... 70 559.264404
71 2020-04-27 sub_region ... 71 559.198975
72 2020-04-28 sub_region ... 72 559.425171
73 2020-04-29 sub_region ... 73 559.358643
74 2020-04-30 sub_region ... 74 559.806274
[21 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/media/plotly/test_plot_12_b8d9f787.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: a9hey38z
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/media/plotly/test_plot_all_13_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181254-a9hey38z/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: a9hey38z INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: xd7cpvm3 with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 5 use_mask: True wandb: Agent Started Run: xd7cpvm3
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/xd7cpvm3 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnhkN2Nwdm0zOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/media/graph/graph_0_summary_ac6126e9.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/media/graph
The running loss is: 20.4773468375206 The number of items in train is: 25 The loss for epoch 0 0.819093873500824
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-summary.json
The running loss is: 19.37137946486473 The number of items in train is: 25 The loss for epoch 1 0.7748551785945892
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-summary.json
The running loss is: 16.317528434097767 The number of items in train is: 25 The loss for epoch 2 0.6527011373639107 The running loss is: 14.260967433452606 The number of items in train is: 25 The loss for epoch 3 0.5704386973381043
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-summary.json
1 The running loss is:
INFO:wandb.wandb_agent:Running runs: ['xd7cpvm3']
14.17140232771635 The number of items in train is: 25 The loss for epoch 4 0.566856093108654
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-summary.json
The running loss is: 14.054600022733212 The number of items in train is: 25 The loss for epoch 5 0.5621840009093284
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-summary.json
The running loss is: 14.07607826590538 The number of items in train is: 25 The loss for epoch 6 0.5630431306362152
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-summary.json
The running loss is: 14.025826334953308 The number of items in train is: 25 The loss for epoch 7 0.5610330533981324 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-summary.json
The running loss is: 13.500305883586407 The number of items in train is: 25 The loss for epoch 8 0.5400122353434562 2 The running loss is: 12.7346723228693 The number of items in train is: 25 The loss for epoch 9 0.509386892914772 3 Stopping model now Data saved to: 28_May_202006_13PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-summary.json
Data saved to: 28_May_202006_13PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params) INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/config.yaml
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 384.512299
66 2020-04-22 sub_region ... 66 383.916504
67 2020-04-23 sub_region ... 67 384.334045
68 2020-04-24 sub_region ... 68 384.549500
69 2020-04-25 sub_region ... 69 384.431915
70 2020-04-26 sub_region ... 70 384.146332
71 2020-04-27 sub_region ... 71 384.198578
72 2020-04-28 sub_region ... 72 383.738586
73 2020-04-29 sub_region ... 73 384.381348
74 2020-04-30 sub_region ... 74 383.990662
[21 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/media/plotly/test_plot_20_1987c841.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: xd7cpvm3
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/media/plotly/test_plot_all_21_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181311-xd7cpvm3/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: xd7cpvm3 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: lt4usxjg with config: batch_size: 2 forecast_history: 11 lr: 0.01 number_encoder_layers: 3 out_seq_length: 5 use_mask: False wandb: Agent Started Run: lt4usxjg
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/lt4usxjg INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmx0NHVzeGpnOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 11
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 11
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 11
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.01
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 11
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.01
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/media/graph/graph_0_summary_76e907eb.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/media
The running loss is: 20.519908547401428 The number of items in train is: 25 The loss for epoch 0 0.8207963418960571
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-summary.json
The running loss is: 19.348501980304718 The number of items in train is: 25 The loss for epoch 1 0.7739400792121888
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-summary.json
The running loss is: 16.020245015621185 The number of items in train is: 25 The loss for epoch 2 0.6408098006248474 The running loss is: 14.293965496122837 The number of items in train is: 25 The loss for epoch 3 0.5717586198449135
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-summary.json
1 The running loss is: 14.129456043243408 The number of items in train is: 25 The loss for epoch 4 0.5651782417297363
INFO:wandb.wandb_agent:Running runs: ['lt4usxjg']
2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-summary.json
The running loss is: 13.794427633285522 The number of items in train is: 25 The loss for epoch 5 0.5517771053314209
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-summary.json
The running loss is: 13.02604092657566 The number of items in train is: 25 The loss for epoch 6 0.5210416370630264 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-summary.json
The running loss is: 9.591085441410542 The number of items in train is: 25 The loss for epoch 7 0.3836434176564217 2 The running loss is: 7.496750168502331 The number of items in train is: 25 The loss for epoch 8 0.29987000674009323 3 Stopping model now Data saved to: 28_May_202006_13PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-summary.json
Data saved to: 28_May_202006_13PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 11, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10])
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/config.yaml
test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 388.992981
66 2020-04-22 sub_region ... 66 387.531372
67 2020-04-23 sub_region ... 67 388.493744
68 2020-04-24 sub_region ... 68 388.772583
69 2020-04-25 sub_region ... 69 388.547852
70 2020-04-26 sub_region ... 70 387.934692
71 2020-04-27 sub_region ... 71 388.074402
72 2020-04-28 sub_region ... 72 387.295563
73 2020-04-29 sub_region ... 73 388.444427
74 2020-04-30 sub_region ... 74 387.736328
[21 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/media/plotly/test_plot_18_166901d4.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: lt4usxjg
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/media/plotly/test_plot_all_19_6655637c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181335-lt4usxjg/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: lt4usxjg INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: d6ex9o8g with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 1 use_mask: True wandb: Agent Started Run: d6ex9o8g
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/d6ex9o8g INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmQ2ZXg5bzhnOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
13.29942782886792
The number of items in train is:
25
The loss for epoch 0
0.5319771131547167
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/media/graph/graph_0_summary_12bf24e1.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/media/graph
The running loss is: 15.958505789749324 The number of items in train is: 25 The loss for epoch 1 0.638340231589973 The running loss is: 9.610189565457404 The number of items in train is: 25 The loss for epoch 2 0.38440758261829616
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-history.jsonl
The running loss is: 8.331762389920186 The number of items in train is: 25 The loss for epoch 3 0.33327049559680744 1 The running loss is: 6.60916328427993 The number of items in train is: 25 The loss for epoch 4 0.2643665313711972 2 The running loss is: 6.482465127395699 The number of items in train is: 25 The loss for epoch 5 0.25929860509582797
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-history.jsonl
The running loss is: 7.1515498894732445 The number of items in train is: 25 The loss for epoch 6 0.2860619955789298 The running loss is: 6.881212463777047 The number of items in train is: 25 The loss for epoch 7 0.27524849855108185 The running loss is: 7.631635777652264 The number of items in train is: 25 The loss for epoch 8 0.30526543110609056 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-history.jsonl
The running loss is: 6.2557811347651295 The number of items in train is: 25 The loss for epoch 9 0.2502312453906052 Data saved to: 28_May_202006_13PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_13PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['d6ex9o8g'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 801.962341
66 2020-04-22 sub_region ... 66 824.838013
67 2020-04-23 sub_region ... 67 775.098450
68 2020-04-24 sub_region ... 68 820.728394
69 2020-04-25 sub_region ... 69 858.230591
70 2020-04-26 sub_region ... 70 790.655151
71 2020-04-27 sub_region ... 71 957.766296
72 2020-04-28 sub_region ... 72 793.469360
73 2020-04-29 sub_region ... 73 769.994263
74 2020-04-30 sub_region ... 74 771.788574
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/media/plotly/test_plot_20_302bf663.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: d6ex9o8g
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181352-d6ex9o8g/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: d6ex9o8g INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 3igvo891 with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 1 use_mask: False wandb: Agent Started Run: 3igvo891
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/3igvo891 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjNpZ3ZvODkxOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
13.510150267975405
The number of items in train is:
25
The loss for epoch 0
0.5404060107190162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/media/graph/graph_0_summary_9cedcb3c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/media/graph
The running loss is: 16.415556019172072 The number of items in train is: 25 The loss for epoch 1 0.6566222407668829 1 The running loss is: 8.791841979138553 The number of items in train is: 25 The loss for epoch 2 0.35167367916554215 The running loss is: 9.764055475010537 The number of items in train is: 25 The loss for epoch 3 0.3905622190004215
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-history.jsonl
The running loss is: 7.065425713779405 The number of items in train is: 25 The loss for epoch 4 0.2826170285511762 1 The running loss is: 6.248815768616623 The number of items in train is: 25 The loss for epoch 5 0.24995263074466492 The running loss is: 6.674299073871225 The number of items in train is: 25 The loss for epoch 6 0.266971962954849
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-history.jsonl
1 The running loss is: 7.030096196685918 The number of items in train is: 25 The loss for epoch 7 0.2812038478674367 The running loss is: 8.11010273359716 The number of items in train is: 25 The loss for epoch 8 0.32440410934388636 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-history.jsonl
The running loss is: 8.296146783512086 The number of items in train is: 25 The loss for epoch 9 0.33184587134048343 Data saved to: 28_May_202006_14PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_14PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['3igvo891']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 725.814941
66 2020-04-22 sub_region ... 66 739.527832
67 2020-04-23 sub_region ... 67 716.998169
68 2020-04-24 sub_region ... 68 738.462158
69 2020-04-25 sub_region ... 69 756.830688
70 2020-04-26 sub_region ... 70 728.941650
71 2020-04-27 sub_region ... 71 807.169800
72 2020-04-28 sub_region ... 72 725.725586
73 2020-04-29 sub_region ... 73 714.828308
74 2020-04-30 sub_region ... 74 716.417358
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 3igvo891
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/media/plotly/test_plot_20_f6cc8bc8.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181410-3igvo891/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 3igvo891 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 2ptjpclj with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 2 use_mask: True wandb: Agent Started Run: 2ptjpclj
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/2ptjpclj INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjJwdGpwY2xqOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
19.00246888399124
The number of items in train is:
25
The loss for epoch 0
0.7600987553596497
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/media/graph/graph_0_summary_6986d7e9.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/media
The running loss is: 16.036757367663085 The number of items in train is: 25 The loss for epoch 1 0.6414702947065234 1 The running loss is: 10.140504654031247 The number of items in train is: 25 The loss for epoch 2 0.4056201861612499 The running loss is: 11.365801957435906 The number of items in train is: 25 The loss for epoch 3 0.45463207829743624
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-history.jsonl
1 The running loss is: 15.48135045915842 The number of items in train is: 25 The loss for epoch 4 0.6192540183663369 2 The running loss is: 8.226837230846286 The number of items in train is: 25 The loss for epoch 5 0.32907348923385144 The running loss is: 7.584667568095028 The number of items in train is: 25 The loss for epoch 6 0.30338670272380114
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-summary.json
1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-history.jsonl
The running loss is: 8.201896103098989 The number of items in train is: 25 The loss for epoch 7 0.3280758441239595 2 The running loss is: 8.102215321734548 The number of items in train is: 25 The loss for epoch 8 0.3240886128693819 The running loss is: 8.849566355347633 The number of items in train is: 25 The loss for epoch 9 0.35398265421390535
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-history.jsonl
1 Data saved to: 28_May_202006_14PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_14PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['2ptjpclj']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/config.yaml /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 721.418884
66 2020-04-22 sub_region ... 66 729.418823
67 2020-04-23 sub_region ... 67 704.071716
68 2020-04-24 sub_region ... 68 698.017456
69 2020-04-25 sub_region ... 69 721.563843
70 2020-04-26 sub_region ... 70 702.487244
71 2020-04-27 sub_region ... 71 729.263550
72 2020-04-28 sub_region ... 72 726.579834
73 2020-04-29 sub_region ... 73 722.222656
74 2020-04-30 sub_region ... 74 718.942139
[25 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/media/plotly/test_plot_20_359c7a7b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 2ptjpclj
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181422-2ptjpclj/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 2ptjpclj INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: jcdp0taa with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 2 use_mask: False wandb: Agent Started Run: jcdp0taa
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/jcdp0taa INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmpjZHAwdGFhOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
18.724818095564842
The number of items in train is:
25
The loss for epoch 0
0.7489927238225937
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/media/graph/graph_0_summary_05e489c0.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/media/graph
The running loss is: 15.727116253226995 The number of items in train is: 25 The loss for epoch 1 0.6290846501290798 1 The running loss is: 8.56138514901977 The number of items in train is: 25 The loss for epoch 2 0.3424554059607908 The running loss is: 10.345093304291368 The number of items in train is: 25 The loss for epoch 3 0.4138037321716547
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-summary.json
1 The running loss is: 10.967490065842867 The number of items in train is: 25 The loss for epoch 4 0.43869960263371466 The running loss is: 7.611721908673644 The number of items in train is: 25 The loss for epoch 5 0.30446887634694575 The running loss is: 7.797869371250272 The number of items in train is: 25 The loss for epoch 6 0.31191477485001085
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-summary.json
The running loss is: 7.795289715752006 The number of items in train is: 25 The loss for epoch 7 0.31181158863008024 The running loss is: 6.245977671816945 The number of items in train is: 25 The loss for epoch 8 0.2498391068726778 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-summary.json
The running loss is: 6.67815606854856 The number of items in train is: 25 The loss for epoch 9 0.2671262427419424 Data saved to: 28_May_202006_14PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_14PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['jcdp0taa'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 648.033325
66 2020-04-22 sub_region ... 66 696.399780
67 2020-04-23 sub_region ... 67 633.036682
68 2020-04-24 sub_region ... 68 599.434021
69 2020-04-25 sub_region ... 69 679.843323
70 2020-04-26 sub_region ... 70 630.065979
71 2020-04-27 sub_region ... 71 673.871094
72 2020-04-28 sub_region ... 72 681.052856
73 2020-04-29 sub_region ... 73 670.992432
74 2020-04-30 sub_region ... 74 661.747437
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/media/plotly/test_plot_20_c2722b2a.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: jcdp0taa
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181439-jcdp0taa/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: jcdp0taa INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: sfn581er with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 3 use_mask: True wandb: Agent Started Run: sfn581er
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/sfn581er INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnNmbjU4MWVyOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
12.417334482073784
The number of items in train is:
24
The loss for epoch 0
0.5173889367530743
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/media/graph/graph_0_summary_e93bba15.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/media/graph
The running loss is: 14.187367584556341 The number of items in train is: 24 The loss for epoch 1 0.5911403160231808 The running loss is: 8.399394869804382 The number of items in train is: 24 The loss for epoch 2 0.34997478624184925 1 The running loss is: 6.990775217302144 The number of items in train is: 24 The loss for epoch 3 0.29128230072092265
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-summary.json
The running loss is: 8.946204421576113 The number of items in train is: 24 The loss for epoch 4 0.37275851756567135 1 The running loss is: 15.735674642026424 The number of items in train is: 24 The loss for epoch 5 0.6556531100844344 The running loss is: 7.7431636694818735 The number of items in train is: 24 The loss for epoch 6 0.32263181956174475
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-summary.json
1 The running loss is: 4.328124134801328 The number of items in train is: 24 The loss for epoch 7 0.18033850561672202 2 The running loss is: 3.735455541405827 The number of items in train is: 24 The loss for epoch 8 0.15564398089190945
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-summary.json
The running loss is: 3.9662495320662856 The number of items in train is: 24 The loss for epoch 9 0.16526039716942856 1 Data saved to: 28_May_202006_15PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_15PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['sfn581er']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 755.517212
66 2020-04-22 sub_region ... 66 771.916016
67 2020-04-23 sub_region ... 67 760.687378
68 2020-04-24 sub_region ... 68 735.305786
69 2020-04-25 sub_region ... 69 728.676880
70 2020-04-26 sub_region ... 70 742.574097
71 2020-04-27 sub_region ... 71 690.534424
72 2020-04-28 sub_region ... 72 759.479492
73 2020-04-29 sub_region ... 73 761.484863
74 2020-04-30 sub_region ... 74 765.487183
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/media/plotly/test_plot_20_1d2f2943.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/media/plotly
wandb: Agent Finished Run: sfn581er
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181459-sfn581er/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: sfn581er INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: gdvkn2kb with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 3 use_mask: False wandb: Agent Started Run: gdvkn2kb
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/gdvkn2kb INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmdkdmtuMmtiOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
12.737247206270695
The number of items in train is:
24
The loss for epoch 0
0.5307186335946122
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/media/graph/graph_0_summary_b9282d02.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/media/graph
The running loss is: 12.846482746303082 The number of items in train is: 24 The loss for epoch 1 0.5352701144292951 The running loss is: 8.131057146936655 The number of items in train is: 24 The loss for epoch 2 0.3387940477890273 1 The running loss is: 5.895240487996489 The number of items in train is: 24 The loss for epoch 3 0.24563502033318704 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-history.jsonl
The running loss is: 8.398755688220263 The number of items in train is: 24 The loss for epoch 4 0.34994815367584425 The running loss is: 16.694155018776655 The number of items in train is: 24 The loss for epoch 5 0.6955897924490273 The running loss is: 9.139933694154024 The number of items in train is: 24 The loss for epoch 6 0.380830570589751 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-history.jsonl
The running loss is: 5.126268687658012 The number of items in train is: 24 The loss for epoch 7 0.21359452865241715 2 The running loss is: 4.945361414458603 The number of items in train is: 24 The loss for epoch 8 0.2060567256024418 3 Stopping model now Data saved to: 28_May_202006_15PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_15PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 596.721497
66 2020-04-22 sub_region ... 66 611.140503
67 2020-04-23 sub_region ... 67 602.298889
68 2020-04-24 sub_region ... 68 592.272522
69 2020-04-25 sub_region ... 69 590.910889
70 2020-04-26 sub_region ... 70 592.348145
71 2020-04-27 sub_region ... 71 565.893555
72 2020-04-28 sub_region ... 72 597.307617
73 2020-04-29 sub_region ... 73 598.141846
74 2020-04-30 sub_region ... 74 604.241028
[25 rows x 28 columns]
INFO:wandb.wandb_agent:Running runs: ['gdvkn2kb']
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/media/plotly/test_plot_18_b0d51432.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: gdvkn2kb
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/media/plotly/test_plot_all_19_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181512-gdvkn2kb/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: gdvkn2kb INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: fqkznitg with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 4 use_mask: True wandb: Agent Started Run: fqkznitg
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/fqkznitg INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmZxa3puaXRnOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
15.76520562171936
The number of items in train is:
24
The loss for epoch 0
0.65688356757164
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/media/graph/graph_0_summary_e7bd39af.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/media
The running loss is: 15.477864772081375 The number of items in train is: 24 The loss for epoch 1 0.6449110321700573 1 The running loss is: 12.414291091263294 The number of items in train is: 24 The loss for epoch 2 0.5172621288026372 2 The running loss is: 7.5504014934413135 The number of items in train is: 24 The loss for epoch 3 0.3146000622267214
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-summary.json
The running loss is: 12.823520310223103 The number of items in train is: 24 The loss for epoch 4 0.5343133462592959 1 The running loss is: 6.759511549025774 The number of items in train is: 24 The loss for epoch 5 0.2816463145427406 The running loss is: 6.30719225294888 The number of items in train is: 24 The loss for epoch 6 0.26279967720620334 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-summary.json
The running loss is: 6.85306416079402 The number of items in train is: 24 The loss for epoch 7 0.28554434003308415 2 The running loss is: 6.9310750886797905 The number of items in train is: 24 The loss for epoch 8 0.2887947953616579 The running loss is: 8.455319091677666 The number of items in train is: 24 The loss for epoch 9 0.3523049621532361 1 Data saved to: 28_May_202006_15PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-summary.json
Data saved to: 28_May_202006_15PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['fqkznitg']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 779.131897
66 2020-04-22 sub_region ... 66 792.555298
67 2020-04-23 sub_region ... 67 780.039795
68 2020-04-24 sub_region ... 68 769.559204
69 2020-04-25 sub_region ... 69 768.763855
70 2020-04-26 sub_region ... 70 775.261108
71 2020-04-27 sub_region ... 71 753.575806
72 2020-04-28 sub_region ... 72 784.202087
73 2020-04-29 sub_region ... 73 783.728699
74 2020-04-30 sub_region ... 74 785.233765
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/media/plotly/test_plot_20_f7c7c22f.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/media/plotly
wandb: Agent Finished Run: fqkznitg
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181524-fqkznitg/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: fqkznitg INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 8svfqbv0 with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 4 use_mask: False wandb: Agent Started Run: 8svfqbv0
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/8svfqbv0 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjhzdmZxYnYwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
15.728306412696838
The number of items in train is:
24
The loss for epoch 0
0.655346100529035
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/media/graph/graph_0_summary_2010ea22.graph.json
The running loss is: 13.534287229180336 The number of items in train is: 24 The loss for epoch 1 0.5639286345491806
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/media/graph
1 The running loss is: 10.27218734100461 The number of items in train is: 24 The loss for epoch 2 0.4280078058751921 2 The running loss is: 5.5147014474496245 The number of items in train is: 24 The loss for epoch 3 0.22977922697706768
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/wandb-history.jsonl
The running loss is: 9.376355066895485 The number of items in train is: 24 The loss for epoch 4 0.3906814611206452 1 The running loss is: 6.485733915120363 The number of items in train is: 24 The loss for epoch 5 0.27023891313001513 2 The running loss is: 7.979923363775015 The number of items in train is: 24 The loss for epoch 6 0.33249680682395893
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/wandb-history.jsonl
The running loss is: 8.143040359020233 The number of items in train is: 24 The loss for epoch 7 0.33929334829250973 1 The running loss is: 5.536451553925872 The number of items in train is: 24 The loss for epoch 8 0.230685481413578 2 The running loss is: 4.550636069383472 The number of items in train is: 24 The loss for epoch 9 0.18960983622431135 Data saved to: 28_May_202006_15PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/wandb-history.jsonl
Data saved to: 28_May_202006_15PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
INFO:wandb.wandb_agent:Running runs: ['8svfqbv0'] /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 756.924194
66 2020-04-22 sub_region ... 66 788.874878
67 2020-04-23 sub_region ... 67 767.624146
68 2020-04-24 sub_region ... 68 745.193481
69 2020-04-25 sub_region ... 69 751.710815
70 2020-04-26 sub_region ... 70 755.871094
71 2020-04-27 sub_region ... 71 710.098877
72 2020-04-28 sub_region ... 72 772.029785
73 2020-04-29 sub_region ... 73 776.352173
74 2020-04-30 sub_region ... 74 783.972778
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 8svfqbv0
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/media/plotly/test_plot_20_75431aa1.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181538-8svfqbv0/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 8svfqbv0 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: myhifaua with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 5 use_mask: True wandb: Agent Started Run: myhifaua
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/myhifaua INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-metadata.json INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm15aGlmYXVhOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
12.287862300872803
The number of items in train is:
23
The loss for epoch 0
0.5342548826466436
The running loss is:
13.926748000085354
The number of items in train is:
23
The loss for epoch 1
0.6055107826124067
1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/media/graph/graph_0_summary_0827234d.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/media/graph
The running loss is: 11.354672580957413 The number of items in train is: 23 The loss for epoch 2 0.4936814165633658 2 The running loss is: 9.702557804062963 The number of items in train is: 23 The loss for epoch 3 0.4218503393070853 The running loss is: 11.627340912818909 The number of items in train is: 23 The loss for epoch 4 0.5055365614269091 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-history.jsonl
The running loss is: 6.340536650270224 The number of items in train is: 23 The loss for epoch 5 0.275675506533488 2 The running loss is: 10.078678345307708 The number of items in train is: 23 The loss for epoch 6 0.4382034063177264
INFO:wandb.wandb_agent:Running runs: ['myhifaua']
The running loss is: 8.461791813373566 The number of items in train is: 23 The loss for epoch 7 0.3679039918858072 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-history.jsonl
The running loss is: 4.9823848977684975 The number of items in train is: 23 The loss for epoch 8 0.21662543033776077 2 The running loss is: 3.317102179862559 The number of items in train is: 23 The loss for epoch 9 0.14422183390706778 Data saved to: 28_May_202006_15PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_15PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 835.124390
66 2020-04-22 sub_region ... 66 840.772278
67 2020-04-23 sub_region ... 67 832.618164
68 2020-04-24 sub_region ... 68 835.653015
69 2020-04-25 sub_region ... 69 821.924072
70 2020-04-26 sub_region ... 70 833.929749
71 2020-04-27 sub_region ... 71 823.226929
72 2020-04-28 sub_region ... 72 831.859680
73 2020-04-29 sub_region ... 73 828.141541
74 2020-04-30 sub_region ... 74 831.716797
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/media/plotly/test_plot_20_7a7a1d5c.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: myhifaua
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181550-myhifaua/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: myhifaua INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: a9pg6ijd with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 1 out_seq_length: 5 use_mask: False wandb: Agent Started Run: a9pg6ijd
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/a9pg6ijd INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmE5cGc2aWpkOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/config.yaml
The running loss is: 12.292471200227737 The number of items in train is: 23 The loss for epoch 0 0.534455269575119
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/media/graph/graph_0_summary_7460a360.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/media
The running loss is: 14.315681867301464 The number of items in train is: 23 The loss for epoch 1 0.6224209507522376 1 The running loss is: 10.289977289736271 The number of items in train is: 23 The loss for epoch 2 0.44739031694505527 2 The running loss is: 7.302962388843298 The number of items in train is: 23
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-summary.json
The loss for epoch 3
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-history.jsonl
0.3175201038627521 The running loss is: 10.639127224683762 The number of items in train is: 23 The loss for epoch 4 0.462570748899294 The running loss is: 5.083698073402047 The number of items in train is: 23 The loss for epoch 5 0.2210303510174803 1 The running loss is: 4.266114007681608 The number of items in train is: 23 The loss for epoch 6 0.1854832177252873
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-history.jsonl
2 The running loss is: 3.362240646034479 The number of items in train is: 23 The loss for epoch 7 0.14618437591454256 The running loss is: 5.346718622371554 The number of items in train is: 23 The loss for epoch 8 0.2324660270596328 1 The running loss is: 3.976900838315487 The number of items in train is: 23 The loss for epoch 9 0.17290873210067334 Data saved to: 28_May_202006_16PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-history.jsonl INFO:wandb.wandb_agent:Running runs: ['a9pg6ijd']
Data saved to: 28_May_202006_16PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 773.226318
66 2020-04-22 sub_region ... 66 785.933838
67 2020-04-23 sub_region ... 67 776.867065
68 2020-04-24 sub_region ... 68 779.012268
69 2020-04-25 sub_region ... 69 770.271240
70 2020-04-26 sub_region ... 70 779.572510
71 2020-04-27 sub_region ... 71 765.994873
72 2020-04-28 sub_region ... 72 776.860962
73 2020-04-29 sub_region ... 73 774.669739
74 2020-04-30 sub_region ... 74 777.345093
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/media/plotly/test_plot_20_b982c2b4.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: a9pg6ijd
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181607-a9pg6ijd/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: a9pg6ijd INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 4r6vcmds with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 1 use_mask: True wandb: Agent Started Run: 4r6vcmds
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/4r6vcmds INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjRyNnZjbWRzOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/media/graph/graph_0_summary_ef8144b2.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/media/graph
The running loss is: 10.145409000921063 The number of items in train is: 25 The loss for epoch 0 0.4058163600368425 The running loss is: 25.701497849076986 The number of items in train is: 25 The loss for epoch 1 1.0280599139630795
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-summary.json
The running loss is: 20.272816001903266 The number of items in train is: 25 The loss for epoch 2 0.8109126400761306 The running loss is: 19.751347083598375 The number of items in train is: 25 The loss for epoch 3 0.790053883343935
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-summary.json
The running loss is: 18.793059289455414 The number of items in train is: 25 The loss for epoch 4 0.7517223715782165 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-summary.json
The running loss is: 16.612464647740126 The number of items in train is: 25 The loss for epoch 5 0.664498585909605 2 The running loss is: 9.692850414896384 The number of items in train is: 25 The loss for epoch 6 0.38771401659585536
INFO:wandb.wandb_agent:Running runs: ['4r6vcmds'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-summary.json
The running loss is: 19.73813858255744 The number of items in train is: 25 The loss for epoch 7 0.7895255433022976 The running loss is: 12.20803095260635 The number of items in train is: 25 The loss for epoch 8 0.488321238104254 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-summary.json
The running loss is: 11.476102620712481 The number of items in train is: 25 The loss for epoch 9 0.4590441048284993 2 Data saved to: 28_May_202006_16PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_16PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 801.515869
66 2020-04-22 sub_region ... 66 801.592773
67 2020-04-23 sub_region ... 67 801.521545
68 2020-04-24 sub_region ... 68 801.597168
69 2020-04-25 sub_region ... 69 801.738037
70 2020-04-26 sub_region ... 70 801.605835
71 2020-04-27 sub_region ... 71 801.837097
72 2020-04-28 sub_region ... 72 801.528931
73 2020-04-29 sub_region ... 73 801.504639
74 2020-04-30 sub_region ... 74 801.555054
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/media/plotly/test_plot_20_fcc89d57.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 4r6vcmds
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181624-4r6vcmds/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 4r6vcmds INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: cy2kokyn with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 1 use_mask: False wandb: Agent Started Run: cy2kokyn
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/cy2kokyn INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmN5Mmtva3luOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/media/graph/graph_0_summary_5bda3603.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/media
The running loss is: 10.096708084311103 The number of items in train is: 25 The loss for epoch 0 0.40386832337244416 The running loss is: 25.173049704171717 The number of items in train is: 25 The loss for epoch 1 1.0069219881668687
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-history.jsonl
The running loss is: 20.567884587217122 The number of items in train is: 25 The loss for epoch 2 0.8227153834886849 The running loss is: 16.932578061707318 The number of items in train is: 25 The loss for epoch 3 0.6773031224682927
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-history.jsonl
1 The running loss is: 11.519542964175344 The number of items in train is: 25 The loss for epoch 4 0.4607817185670137 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-history.jsonl
The running loss is: 8.808077305089682 The number of items in train is: 25 The loss for epoch 5 0.3523230922035873 The running loss is: 11.38830050965771 The number of items in train is: 25 The loss for epoch 6 0.4555320203863084 1
INFO:wandb.wandb_agent:Running runs: ['cy2kokyn'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-history.jsonl
The running loss is: 7.139500390738249 The number of items in train is: 25 The loss for epoch 7 0.28558001562952995 2 The running loss is: 9.381748400162905 The number of items in train is: 25 The loss for epoch 8 0.3752699360065162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-history.jsonl
The running loss is: 10.521757567301393 The number of items in train is: 25 The loss for epoch 9 0.4208703026920557 Data saved to: 28_May_202006_16PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_16PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 761.296265
66 2020-04-22 sub_region ... 66 761.453735
67 2020-04-23 sub_region ... 67 761.306885
68 2020-04-24 sub_region ... 68 761.469482
69 2020-04-25 sub_region ... 69 761.735107
70 2020-04-26 sub_region ... 70 761.236694
71 2020-04-27 sub_region ... 71 761.615662
72 2020-04-28 sub_region ... 72 760.781616
73 2020-04-29 sub_region ... 73 760.612366
74 2020-04-30 sub_region ... 74 760.868896
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/media/plotly/test_plot_20_b3c2d6a3.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: cy2kokyn
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181641-cy2kokyn/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: cy2kokyn INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: p1uy2kwb with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 2 use_mask: True wandb: Agent Started Run: p1uy2kwb
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/p1uy2kwb INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnAxdXkya3diOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/media/graph/graph_0_summary_085426ee.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/media
The running loss is: 20.584014743566513 The number of items in train is: 25 The loss for epoch 0 0.8233605897426606 The running loss is: 17.932957649230957 The number of items in train is: 25 The loss for epoch 1 0.7173183059692383 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-history.jsonl
The running loss is: 15.548911690013483 The number of items in train is: 25 The loss for epoch 2 0.6219564676005394 The running loss is: 14.730191620066762 The number of items in train is: 25 The loss for epoch 3 0.5892076648026705
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-history.jsonl
The running loss is: 15.559959415346384 The number of items in train is: 25 The loss for epoch 4 0.6223983766138553 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-history.jsonl
The running loss is: 11.396017402410507 The number of items in train is: 25 The loss for epoch 5 0.4558406960964203 2 The running loss is: 9.876246387138963 The number of items in train is: 25 The loss for epoch 6 0.3950498554855585 3 Stopping model now Data saved to: 28_May_202006_17PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.wandb_agent:Running runs: ['p1uy2kwb']
Data saved to: 28_May_202006_17PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 505.433014
66 2020-04-22 sub_region ... 66 504.062500
67 2020-04-23 sub_region ... 67 504.517761
68 2020-04-24 sub_region ... 68 503.535156
69 2020-04-25 sub_region ... 69 506.940369
70 2020-04-26 sub_region ... 70 505.692200
71 2020-04-27 sub_region ... 71 506.551117
72 2020-04-28 sub_region ... 72 507.229309
73 2020-04-29 sub_region ... 73 507.427246
74 2020-04-30 sub_region ... 74 507.579987
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/media/plotly/test_plot_14_c7166984.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: p1uy2kwb
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/media/plotly/test_plot_all_15_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181658-p1uy2kwb/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: p1uy2kwb INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: d2rvnib6 with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 2 use_mask: False wandb: Agent Started Run: d2rvnib6
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/d2rvnib6 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmQycnZuaWI2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/media/graph/graph_0_summary_1aff1725.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/media/graph
The running loss is: 20.51875665783882 The number of items in train is: 25 The loss for epoch 0 0.8207502663135529 The running loss is: 18.745911203324795 The number of items in train is: 25 The loss for epoch 1 0.7498364481329918 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-summary.json
The running loss is: 14.171599298715591 The number of items in train is: 25 The loss for epoch 2 0.5668639719486237 2 The running loss is: 10.58555408520624 The number of items in train is: 25 The loss for epoch 3 0.4234221634082496
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-summary.json
The running loss is: 17.37247646227479 The number of items in train is: 25 The loss for epoch 4 0.6948990584909915
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-summary.json
The running loss is: 10.633864669129252 The number of items in train is: 25 The loss for epoch 5 0.4253545867651701 1 The running loss is: 9.219570185989141 The number of items in train is: 25 The loss for epoch 6 0.36878280743956565 2
INFO:wandb.wandb_agent:Running runs: ['d2rvnib6'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-summary.json
The running loss is: 8.460247997194529 The number of items in train is: 25 The loss for epoch 7 0.33840991988778113 3 Stopping model now Data saved to: 28_May_202006_17PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_17PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 684.219727
66 2020-04-22 sub_region ... 66 684.323364
67 2020-04-23 sub_region ... 67 684.080750
68 2020-04-24 sub_region ... 68 683.885254
69 2020-04-25 sub_region ... 69 685.422363
70 2020-04-26 sub_region ... 70 684.204346
71 2020-04-27 sub_region ... 71 685.634094
72 2020-04-28 sub_region ... 72 684.494141
73 2020-04-29 sub_region ... 73 684.385315
74 2020-04-30 sub_region ... 74 684.536133
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/media/plotly/test_plot_16_e1926f30.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: d2rvnib6
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/media/plotly/test_plot_all_17_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181715-d2rvnib6/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: d2rvnib6 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: gg4ud0e1 with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 3 use_mask: True wandb: Agent Started Run: gg4ud0e1
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/gg4ud0e1 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmdnNHVkMGUxOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/media/graph/graph_0_summary_fbd696d7.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/media
The running loss is: 18.772414907813072 The number of items in train is: 24 The loss for epoch 0 0.7821839544922113 The running loss is: 19.575059436261654 The number of items in train is: 24 The loss for epoch 1 0.8156274765109023
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-history.jsonl
The running loss is: 15.316046819090843 The number of items in train is: 24 The loss for epoch 2 0.6381686174621185 The running loss is: 11.947806634008884 The number of items in train is: 24 The loss for epoch 3 0.49782527641703683
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-history.jsonl
The running loss is: 10.35117731243372 The number of items in train is: 24 The loss for epoch 4 0.43129905468473834 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-history.jsonl
The running loss is: 11.100783228874207 The number of items in train is: 24 The loss for epoch 5 0.4625326345364253 2 The running loss is: 11.392054236494005 The number of items in train is: 24 The loss for epoch 6 0.4746689265205835
INFO:wandb.wandb_agent:Running runs: ['gg4ud0e1'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-history.jsonl
The running loss is: 5.173449165187776 The number of items in train is: 24 The loss for epoch 7 0.215560381882824 1 The running loss is: 9.032942300196737 The number of items in train is: 24 The loss for epoch 8 0.3763725958415307
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-history.jsonl
The running loss is: 19.382783502340317 The number of items in train is: 24 The loss for epoch 9 0.8076159792641798 Data saved to: 28_May_202006_17PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_17PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 515.919434
66 2020-04-22 sub_region ... 66 516.543030
67 2020-04-23 sub_region ... 67 515.078186
68 2020-04-24 sub_region ... 68 516.592896
69 2020-04-25 sub_region ... 69 515.011414
70 2020-04-26 sub_region ... 70 514.803894
71 2020-04-27 sub_region ... 71 516.728943
72 2020-04-28 sub_region ... 72 515.620483
73 2020-04-29 sub_region ... 73 515.851013
74 2020-04-30 sub_region ... 74 514.158508
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/media/plotly/test_plot_20_14ad17d9.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: gg4ud0e1
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181733-gg4ud0e1/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: gg4ud0e1 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 0xnq5rx0 with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 3 use_mask: False wandb: Agent Started Run: 0xnq5rx0
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/0xnq5rx0 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjB4bnE1cngwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/media/graph/graph_0_summary_45195203.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/media
The running loss is: 18.797405660152435 The number of items in train is: 24 The loss for epoch 0 0.7832252358396848 The running loss is: 19.402401946485043 The number of items in train is: 24 The loss for epoch 1 0.8084334144368768
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-summary.json
The running loss is: 14.805835217237473 The number of items in train is: 24 The loss for epoch 2 0.616909800718228 The running loss is: 9.891540326178074 The number of items in train is: 24 The loss for epoch 3 0.4121475135907531 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-summary.json
The running loss is: 5.6078918017446995 The number of items in train is: 24 The loss for epoch 4 0.23366215840602914 2 The running loss is: 4.041457449435256 The number of items in train is: 24 The loss for epoch 5
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-history.jsonl
0.16839406039313567
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-summary.json
3 Stopping model now Data saved to: 28_May_202006_17PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_17PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['0xnq5rx0']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9])
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/config.yaml
Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-summary.json /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 378.001099
66 2020-04-22 sub_region ... 66 379.296204
67 2020-04-23 sub_region ... 67 378.866272
68 2020-04-24 sub_region ... 68 383.898071
69 2020-04-25 sub_region ... 69 369.884155
70 2020-04-26 sub_region ... 70 381.321381
71 2020-04-27 sub_region ... 71 371.926025
72 2020-04-28 sub_region ... 72 378.183258
73 2020-04-29 sub_region ... 73 379.500549
74 2020-04-30 sub_region ... 74 378.237946
[25 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/media/plotly/test_plot_12_9498e4f6.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 0xnq5rx0
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/media/plotly/test_plot_all_13_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181750-0xnq5rx0/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 0xnq5rx0 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: juzsqvdl with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 4 use_mask: True wandb: Agent Started Run: juzsqvdl
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/juzsqvdl INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmp1enNxdmRsOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/media/graph/graph_0_summary_28380503.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/media/graph
The running loss is: 21.451373293995857
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/media
The number of items in train is: 24 The loss for epoch 0 0.8938072205831608 The running loss is: 18.959119454026222 The number of items in train is: 24 The loss for epoch 1 0.7899633105844259
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-history.jsonl
The running loss is: 16.811535373330116 The number of items in train is: 24 The loss for epoch 2 0.7004806405554215 The running loss is: 15.114799819886684 The number of items in train is: 24 The loss for epoch 3 0.6297833258286119 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-history.jsonl
The running loss is: 13.837378360331059 The number of items in train is: 24 The loss for epoch 4 0.5765574316804608 2 The running loss is: 10.938715167343616 The number of items in train is: 24 The loss for epoch 5 0.45577979863931734
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-history.jsonl
3 Stopping model now Data saved to: 28_May_202006_18PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_18PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['juzsqvdl']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 379.669922
66 2020-04-22 sub_region ... 66 381.567261
67 2020-04-23 sub_region ... 67 381.375244
68 2020-04-24 sub_region ... 68 386.184875
69 2020-04-25 sub_region ... 69 368.961060
70 2020-04-26 sub_region ... 70 383.574829
71 2020-04-27 sub_region ... 71 370.266907
72 2020-04-28 sub_region ... 72 379.182159
73 2020-04-29 sub_region ... 73 380.702667
74 2020-04-30 sub_region ... 74 380.433716
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-history.jsonl
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/media/plotly/test_plot_12_ab9062b5.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/media/plotly
wandb: Agent Finished Run: juzsqvdl
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/media/plotly/test_plot_all_13_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181802-juzsqvdl/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: juzsqvdl INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 62jkcm81 with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 4 use_mask: False wandb: Agent Started Run: 62jkcm81
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/62jkcm81 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjYyamtjbTgxOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/config.yaml
The running loss is: 21.45643301308155 The number of items in train is: 24 The loss for epoch 0 0.8940180422117313
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/media/graph/graph_0_summary_10dd7fe2.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/media
The running loss is: 18.878746390342712 The number of items in train is: 24 The loss for epoch 1 0.7866144329309464
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-summary.json
The running loss is: 16.909177042543888 The number of items in train is: 24 The loss for epoch 2 0.7045490434393287 The running loss is: 14.992017075419426 The number of items in train is: 24 The loss for epoch 3 0.6246673781424761 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-summary.json
The running loss is: 13.057133480906487 The number of items in train is: 24 The loss for epoch 4 0.5440472283711036 2 The running loss is: 7.168162252753973 The number of items in train is: 24 The loss for epoch 5
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-summary.json
0.2986734271980822 3 Stopping model now Data saved to: 28_May_202006_18PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_18PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['62jkcm81'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 381.720642
66 2020-04-22 sub_region ... 66 383.487915
67 2020-04-23 sub_region ... 67 382.618774
68 2020-04-24 sub_region ... 68 386.354279
69 2020-04-25 sub_region ... 69 371.824524
70 2020-04-26 sub_region ... 70 384.303436
71 2020-04-27 sub_region ... 71 373.255554
72 2020-04-28 sub_region ... 72 381.710022
73 2020-04-29 sub_region ... 73 383.322937
74 2020-04-30 sub_region ... 74 382.421906
[25 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/media/plotly/test_plot_12_d4e101e3.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 62jkcm81
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/media/plotly/test_plot_all_13_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181817-62jkcm81/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 62jkcm81 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: qzc31mty with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 5 use_mask: True wandb: Agent Started Run: qzc31mty
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/qzc31mty INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnF6YzMxbXR5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/config.yaml
The running loss is:
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-history.jsonl
17.729351297020912 The number of items in train is:
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-summary.json
23 The loss for epoch 0 0.7708413607400396
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/media/graph/graph_0_summary_59caac40.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/media/graph
The running loss is: 17.55451713502407 The number of items in train is: 23 The loss for epoch 1 0.7632398754358292
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-history.jsonl
The running loss is: 13.769226357340813 The number of items in train is: 23 The loss for epoch 2 0.5986620155365571 1 The running loss is: 10.689910009503365 The number of items in train is: 23 The loss for epoch 3 0.46477869606536365 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-history.jsonl
The running loss is: 10.719441205263138 The number of items in train is: 23 The loss for epoch 4 0.4660626610983973 The running loss is: 11.102143481373787 The number of items in train is: 23 The loss for epoch 5 0.48270189049451245 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-history.jsonl
The running loss is: 8.552160829305649 The number of items in train is: 23 The loss for epoch 6 0.3718330795350282
INFO:wandb.wandb_agent:Running runs: ['qzc31mty']
The running loss is: 8.911982571706176 The number of items in train is: 23 The loss for epoch 7 0.3874775031176598 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-history.jsonl
The running loss is: 10.839222326874733 The number of items in train is: 23 The loss for epoch 8 0.4712705359510753 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-summary.json
The running loss is: 6.471042025834322
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-history.jsonl
The number of items in train is: 23 The loss for epoch 9 0.28134965329714445 Data saved to: 28_May_202006_18PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_18PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-history.jsonl
CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 682.042175
66 2020-04-22 sub_region ... 66 684.554626
67 2020-04-23 sub_region ... 67 676.001160
68 2020-04-24 sub_region ... 68 688.336792
69 2020-04-25 sub_region ... 69 666.439087
70 2020-04-26 sub_region ... 70 671.417664
71 2020-04-27 sub_region ... 71 689.934692
72 2020-04-28 sub_region ... 72 674.941956
73 2020-04-29 sub_region ... 73 665.878967
74 2020-04-30 sub_region ... 74 658.025818
[25 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/media/plotly/test_plot_20_790f76c3.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: qzc31mty
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181828-qzc31mty/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: qzc31mty INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: hs2tp4rm with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 2 out_seq_length: 5 use_mask: False wandb: Agent Started Run: hs2tp4rm
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/hs2tp4rm INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmhzMnRwNHJtOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/config.yaml
The running loss is: 17.749317228794098 The number of items in train is: 23 The loss for epoch 0 0.7717094447301782
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/media/graph/graph_0_summary_fc1378dd.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/media/graph
The running loss is: 17.551174700260162 The number of items in train is: 23 The loss for epoch 1 0.7630945521852245
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-summary.json
The running loss is: 13.54475101083517 The number of items in train is: 23 The loss for epoch 2 0.5889022178623987 1 The running loss is: 10.344177260994911 The number of items in train is: 23 The loss for epoch 3 0.44974683743456134 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-summary.json
The running loss is: 9.028607338666916 The number of items in train is: 23 The loss for epoch 4 0.39254814515943115 3 Stopping model now Data saved to: 28_May_202006_18PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_18PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 391.728210
66 2020-04-22 sub_region ... 66 392.369507
67 2020-04-23 sub_region ... 67 392.830872
68 2020-04-24 sub_region ... 68 397.064850
69 2020-04-25 sub_region ... 69 385.320221
70 2020-04-26 sub_region ... 70 393.841858
71 2020-04-27 sub_region ... 71 388.418610
72 2020-04-28 sub_region ... 72 390.187714
73 2020-04-29 sub_region ... 73 390.757416
74 2020-04-30 sub_region ... 74 391.191467
[25 rows x 28 columns]
INFO:wandb.wandb_agent:Running runs: ['hs2tp4rm']
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-history.jsonl
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/media/plotly/test_plot_10_ca77e0e0.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: hs2tp4rm
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/media/plotly/test_plot_all_11_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181846-hs2tp4rm/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: hs2tp4rm INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: crrevwn9 with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 1 use_mask: True wandb: Agent Started Run: crrevwn9
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/crrevwn9 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmNycmV2d245OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/media/graph/graph_0_summary_282b85a8.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/media
The running loss is: 10.765136603731662 The number of items in train is: 25 The loss for epoch 0 0.4306054641492665
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-history.jsonl
The running loss is: 26.137935783714056 The number of items in train is: 25 The loss for epoch 1 1.0455174313485622
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-history.jsonl
The running loss is: 22.011536345118657 The number of items in train is: 25 The loss for epoch 2 0.8804614538047463
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-history.jsonl
The running loss is: 18.903898189775646 The number of items in train is: 25 The loss for epoch 3 0.7561559275910258
INFO:wandb.wandb_agent:Running runs: ['crrevwn9'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-history.jsonl
The running loss is: 18.865029871463776 The number of items in train is: 25 The loss for epoch 4 0.754601194858551 The running loss is: 19.70023591836798 The number of items in train is: 25 The loss for epoch 5 0.7880094367347192
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-history.jsonl
1 The running loss is: 18.773742999881506 The number of items in train is: 25 The loss for epoch 6 0.7509497199952603
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-history.jsonl
The running loss is: 19.47792062163353 The number of items in train is: 25 The loss for epoch 7 0.7791168248653412
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-history.jsonl
The running loss is: 19.640535548329353 The number of items in train is: 25 The loss for epoch 8 0.7856214219331741 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-history.jsonl
The running loss is: 19.046687852591276 The number of items in train is: 25 The loss for epoch 9 0.7618675141036511 Data saved to: 28_May_202006_19PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_19PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 496.016388
66 2020-04-22 sub_region ... 66 496.046783
67 2020-04-23 sub_region ... 67 495.631775
68 2020-04-24 sub_region ... 68 495.853607
69 2020-04-25 sub_region ... 69 496.806030
70 2020-04-26 sub_region ... 70 495.802399
71 2020-04-27 sub_region ... 71 497.801544
72 2020-04-28 sub_region ... 72 496.134644
73 2020-04-29 sub_region ... 73 495.858582
74 2020-04-30 sub_region ... 74 495.841339
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/media/plotly
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/media/plotly/test_plot_20_59b2352c.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: crrevwn9
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181859-crrevwn9/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: crrevwn9 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 73l0xd2r with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 1 use_mask: False wandb: Agent Started Run: 73l0xd2r
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/73l0xd2r INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjczbDB4ZDJyOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/media/graph/graph_0_summary_eaf0769d.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/media/graph
The running loss is: 11.082796600530855 The number of items in train is: 25 The loss for epoch 0 0.44331186402123424
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-summary.json
The running loss is: 26.767839308828115 The number of items in train is: 25 The loss for epoch 1 1.0707135723531247
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-summary.json
The running loss is: 19.91911711357534 The number of items in train is: 25 The loss for epoch 2 0.7967646845430135
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-summary.json
The running loss is: 18.481270607560873 The number of items in train is: 25 The loss for epoch 3 0.7392508243024349 1
INFO:wandb.wandb_agent:Running runs: ['73l0xd2r']
The running loss is: 18.93701122701168 The number of items in train is: 25 The loss for epoch 4 0.7574804490804672
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-summary.json
The running loss is: 19.590259796008468 The number of items in train is: 25 The loss for epoch 5 0.7836103918403388
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-summary.json
The running loss is: 18.897615873254836 The number of items in train is: 25 The loss for epoch 6 0.7559046349301934 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-summary.json
The running loss is: 19.6400615721941 The number of items in train is: 25 The loss for epoch 7 0.785602462887764
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-summary.json
The running loss is: 19.225644499063492 The number of items in train is: 25 The loss for epoch 8 0.7690257799625396 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-summary.json
The running loss is: 19.104803455993533 The number of items in train is: 25 The loss for epoch 9 0.7641921382397413 2 Data saved to: 28_May_202006_19PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_19PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 495.774384
66 2020-04-22 sub_region ... 66 495.760284
67 2020-04-23 sub_region ... 67 495.068512
68 2020-04-24 sub_region ... 68 496.070099
69 2020-04-25 sub_region ... 69 497.589020
70 2020-04-26 sub_region ... 70 495.803619
71 2020-04-27 sub_region ... 71 500.093933
72 2020-04-28 sub_region ... 72 495.899109
73 2020-04-29 sub_region ... 73 495.294128
74 2020-04-30 sub_region ... 74 495.290405
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/media/plotly/test_plot_20_b4b150f6.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 73l0xd2r
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181922-73l0xd2r/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 73l0xd2r INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: e0266c8z with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 2 use_mask: True wandb: Agent Started Run: e0266c8z
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/e0266c8z INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmUwMjY2Yzh6OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/media/graph/graph_0_summary_c2b33b0b.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/media
The running loss is: 22.70665431022644 The number of items in train is: 25 The loss for epoch 0 0.9082661724090576
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-summary.json
The running loss is: 20.07255493104458 The number of items in train is: 25 The loss for epoch 1 0.8029021972417831
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-summary.json
The running loss is: 19.82785291969776 The number of items in train is: 25 The loss for epoch 2 0.7931141167879104 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-summary.json
The running loss is: 21.11555242538452 The number of items in train is: 25 The loss for epoch 3 0.8446220970153808
INFO:wandb.wandb_agent:Running runs: ['e0266c8z'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-summary.json
The running loss is: 19.238652385771275 The number of items in train is: 25 The loss for epoch 4 0.769546095430851 1 The running loss is: 19.169510439038277 The number of items in train is: 25 The loss for epoch 5 0.766780417561531 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-summary.json
The running loss is: 19.454397417604923 The number of items in train is: 25 The loss for epoch 6 0.778175896704197 3 Stopping model now Data saved to: 28_May_202006_19PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-summary.json
Data saved to: 28_May_202006_19PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 386.295502
66 2020-04-22 sub_region ... 66 386.368958
67 2020-04-23 sub_region ... 67 386.170776
68 2020-04-24 sub_region ... 68 386.041290
69 2020-04-25 sub_region ... 69 386.396393
70 2020-04-26 sub_region ... 70 386.061401
71 2020-04-27 sub_region ... 71 386.280853
72 2020-04-28 sub_region ... 72 386.269623
73 2020-04-29 sub_region ... 73 386.305389
74 2020-04-30 sub_region ... 74 386.221985
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/media/plotly/test_plot_14_416120dd.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: e0266c8z
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/media/plotly/test_plot_all_15_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_181945-e0266c8z/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: e0266c8z INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 5bgwtdhu with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 2 use_mask: False wandb: Agent Started Run: 5bgwtdhu
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/5bgwtdhu INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjViZ3d0ZGh1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/media/graph/graph_0_summary_d588c064.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/media
The running loss is: 22.725555673241615 The number of items in train is: 25 The loss for epoch 0 0.9090222269296646
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-history.jsonl
The running loss is: 20.579730972647667 The number of items in train is: 25 The loss for epoch 1 0.8231892389059067
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-history.jsonl
The running loss is: 19.793140269815922 The number of items in train is: 25 The loss for epoch 2 0.7917256107926369 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-history.jsonl
The running loss is: 20.604701928794384 The number of items in train is: 25 The loss for epoch 3 0.8241880771517753
INFO:wandb.wandb_agent:Running runs: ['5bgwtdhu']
The running loss is: 19.242981515824795 The number of items in train is: 25 The loss for epoch 4 0.7697192606329918
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-history.jsonl
1 The running loss is: 19.15913762152195 The number of items in train is: 25 The loss for epoch 5 0.766365504860878 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-history.jsonl
The running loss is: 19.610484287142754 The number of items in train is: 25 The loss for epoch 6 0.7844193714857102 3 Stopping model now Data saved to: 28_May_202006_20PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-history.jsonl
Data saved to: 28_May_202006_20PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 408.607574
66 2020-04-22 sub_region ... 66 408.763275
67 2020-04-23 sub_region ... 67 408.592804
68 2020-04-24 sub_region ... 68 408.707031
69 2020-04-25 sub_region ... 69 408.915588
70 2020-04-26 sub_region ... 70 408.787262
71 2020-04-27 sub_region ... 71 409.198151
72 2020-04-28 sub_region ... 72 408.587250
73 2020-04-29 sub_region ... 73 408.494568
74 2020-04-30 sub_region ... 74 408.603363
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 5bgwtdhu
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/media/plotly/test_plot_14_755375b7.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/media/plotly/test_plot_all_15_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182002-5bgwtdhu/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 5bgwtdhu INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: m0sfcpsb with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 3 use_mask: True wandb: Agent Started Run: m0sfcpsb
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/m0sfcpsb INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm0wc2ZjcHNiOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/media/graph/graph_0_summary_4985a567.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/media/graph
The running loss is: 19.80333824455738 The number of items in train is: 24 The loss for epoch 0 0.8251390935232242
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-summary.json
The running loss is: 22.39603229612112 The number of items in train is: 24 The loss for epoch 1 0.93316801233838
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-summary.json
The running loss is: 14.272497054189444 The number of items in train is: 24 The loss for epoch 2 0.5946873772578934 1 The running loss is: 14.552330110222101 The number of items in train is: 24 The loss for epoch 3 0.6063470879259208
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-summary.json INFO:wandb.wandb_agent:Running runs: ['m0sfcpsb']
The running loss is: 13.988487225025892 The number of items in train is: 24 The loss for epoch 4 0.5828536343760788
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-summary.json
The running loss is: 14.61138915270567 The number of items in train is: 24 The loss for epoch 5 0.6088078813627362 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-summary.json
The running loss is: 13.06570915877819 The number of items in train is: 24 The loss for epoch 6 0.5444045482824246 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-summary.json
The running loss is: 10.021730467677116 The number of items in train is: 24 The loss for epoch 7 0.41757210281987983
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-summary.json
The running loss is: 11.953084118664265 The number of items in train is: 24 The loss for epoch 8 0.49804517161101103 1 The running loss is: 10.495671391487122 The number of items in train is: 24 The loss for epoch 9
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-summary.json
0.4373196413119634 2 Data saved to: 28_May_202006_20PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_20PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 705.102661
66 2020-04-22 sub_region ... 66 705.103027
67 2020-04-23 sub_region ... 67 705.096924
68 2020-04-24 sub_region ... 68 705.107422
69 2020-04-25 sub_region ... 69 705.110229
70 2020-04-26 sub_region ... 70 705.101257
71 2020-04-27 sub_region ... 71 705.131958
72 2020-04-28 sub_region ... 72 705.105957
73 2020-04-29 sub_region ... 73 705.100586
74 2020-04-30 sub_region ... 74 705.097656
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/media/plotly/test_plot_20_2b3d8713.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: m0sfcpsb
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182019-m0sfcpsb/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: m0sfcpsb INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: b4jmjakq with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 3 use_mask: False wandb: Agent Started Run: b4jmjakq
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/b4jmjakq INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmI0am1qYWtxOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/media/graph/graph_0_summary_4978548b.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/media
The running loss is: 19.914355665445328 The number of items in train is: 24 The loss for epoch 0 0.8297648193935553
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-summary.json
The running loss is: 22.440757513046265 The number of items in train is: 24 The loss for epoch 1 0.9350315630435944
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-summary.json
The running loss is: 14.276884399354458 The number of items in train is: 24 The loss for epoch 2 0.5948701833064357 1 The running loss is: 14.543040201067924 The number of items in train is: 24 The loss for epoch 3 0.6059600083778302
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-summary.json INFO:wandb.wandb_agent:Running runs: ['b4jmjakq']
The running loss is: 14.132904570549726 The number of items in train is: 24 The loss for epoch 4 0.5888710237729052
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-summary.json
The running loss is: 13.135652396827936 The number of items in train is: 24 The loss for epoch 5 0.5473188498678306
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-summary.json
The running loss is: 12.404533512890339 The number of items in train is: 24 The loss for epoch 6 0.5168555630370975 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-summary.json
The running loss is: 12.101209737360477 The number of items in train is: 24 The loss for epoch 7 0.5042170723900199
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-summary.json
The running loss is: 14.456808343529701 The number of items in train is: 24 The loss for epoch 8 0.6023670143137375 1 The running loss is: 8.79074577242136 The number of items in train is: 24 The loss for epoch 9 0.36628107385089 2 Data saved to: 28_May_202006_20PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-summary.json
Data saved to: 28_May_202006_20PM_model.pth interpolate should be below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/config.yaml
Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 748.224915
66 2020-04-22 sub_region ... 66 748.223572
67 2020-04-23 sub_region ... 67 748.228333
68 2020-04-24 sub_region ... 68 748.199829
69 2020-04-25 sub_region ... 69 748.202820
70 2020-04-26 sub_region ... 70 748.207642
71 2020-04-27 sub_region ... 71 748.156372
72 2020-04-28 sub_region ... 72 748.215271
73 2020-04-29 sub_region ... 73 748.225708
74 2020-04-30 sub_region ... 74 748.218994
[25 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/media/plotly/test_plot_20_62c51e74.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: b4jmjakq
INFO:wandb.run_manager:shutting down system stats and metadata service
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-events.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/media/plotly/test_plot_all_21_b4a9c644.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/media/plotly
INFO:wandb.run_manager:stopping streaming files and file change observer
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182042-b4jmjakq/wandb-metadata.json
INFO:wandb.wandb_agent:Cleaning up finished run: b4jmjakq
wandb: Network error resolved after 0:00:15.588386, resuming normal operation.
INFO:wandb.wandb_agent:Agent received command: run
INFO:wandb.wandb_agent:Agent starting run with config:
batch_size: 2
forecast_history: 15
lr: 0.001
number_encoder_layers: 3
out_seq_length: 4
use_mask: False
DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: zlpt6bze with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 4 use_mask: False wandb: Agent Started Run: zlpt6bze
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/zlpt6bze INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnpscHQ2YnplOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/media/graph/graph_0_summary_3777feef.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/media/graph
The running loss is: 22.53992810845375 The number of items in train is: 24 The loss for epoch 0 0.939163671185573
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-summary.json
The running loss is: 19.258116833865643 The number of items in train is: 24 The loss for epoch 1 0.8024215347444018
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-summary.json
The running loss is: 16.27618619799614 The number of items in train is: 24 The loss for epoch 2 0.6781744249165058 The running loss is: 15.538168050348759 The number of items in train is: 24 The loss for epoch 3 0.6474236687645316
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-summary.json
The running loss is: 15.75120346993208 The number of items in train is: 24 The loss for epoch 4 0.6563001445805033
INFO:wandb.wandb_agent:Running runs: ['zlpt6bze']
1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-summary.json
The running loss is: 15.720529459416866 The number of items in train is: 24 The loss for epoch 5 0.6550220608090361
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-summary.json
The running loss is: 15.435486644506454 The number of items in train is: 24 The loss for epoch 6 0.6431452768544356
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-summary.json
The running loss is: 15.386884167790413 The number of items in train is: 24 The loss for epoch 7 0.6411201736579338 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-summary.json
The running loss is: 15.28485395014286 The number of items in train is: 24 The loss for epoch 8 0.6368689145892859 The running loss is: 15.743594877421856 The number of items in train is: 24 The loss for epoch 9 0.6559831198925773 1 Data saved to: 28_May_202006_21PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-summary.json
Data saved to: 28_May_202006_21PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 400.785828
66 2020-04-22 sub_region ... 66 401.776733
67 2020-04-23 sub_region ... 67 401.619812
68 2020-04-24 sub_region ... 68 403.111847
69 2020-04-25 sub_region ... 69 400.459259
70 2020-04-26 sub_region ... 70 402.953522
71 2020-04-27 sub_region ... 71 400.775208
72 2020-04-28 sub_region ... 72 400.355286
73 2020-04-29 sub_region ... 73 400.579224
74 2020-04-30 sub_region ... 74 401.702911
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: zlpt6bze
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/media/plotly/test_plot_20_42d1f4aa.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182118-zlpt6bze/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: zlpt6bze INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: gmmqn5n2 with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 5 use_mask: True wandb: Agent Started Run: gmmqn5n2
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/gmmqn5n2 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmdtbXFuNW4yOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/media/graph/graph_0_summary_046522db.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/media
The running loss is: 19.067732363939285 The number of items in train is: 23 The loss for epoch 0 0.8290318419104037
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-summary.json
The running loss is: 18.485696509480476 The number of items in train is: 23 The loss for epoch 1 0.8037259351948033 The running loss is: 13.346832662820816 The number of items in train is: 23 The loss for epoch 2 0.5802970722965572
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-summary.json
The running loss is: 13.152954459190369 The number of items in train is: 23 The loss for epoch 3 0.5718675851821899 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-summary.json
The running loss is: 11.9220210313797 The number of items in train is: 23 The loss for epoch 4 0.5183487404947695
INFO:wandb.wandb_agent:Running runs: ['gmmqn5n2'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-summary.json
The running loss is: 8.841944128274918 The number of items in train is: 23 The loss for epoch 5 0.38443235340325727 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-summary.json
The running loss is: 11.500840276479721 The number of items in train is: 23 The loss for epoch 6 0.5000365337599879 The running loss is: 7.89447258785367 The number of items in train is: 23 The loss for epoch 7 0.3432379386023335 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-summary.json
The running loss is: 7.601438160985708 The number of items in train is: 23 The loss for epoch 8 0.3304973113472047 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-summary.json
The running loss is: 4.348183668218553 The number of items in train is: 23 The loss for epoch 9 0.18905146383558927 Data saved to: 28_May_202006_21PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_21PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 752.961365
66 2020-04-22 sub_region ... 66 752.971558
67 2020-04-23 sub_region ... 67 752.971436
68 2020-04-24 sub_region ... 68 752.982910
69 2020-04-25 sub_region ... 69 753.031128
70 2020-04-26 sub_region ... 70 753.011536
71 2020-04-27 sub_region ... 71 753.077209
72 2020-04-28 sub_region ... 72 752.963745
73 2020-04-29 sub_region ... 73 752.952332
74 2020-04-30 sub_region ... 74 752.973022
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: gmmqn5n2
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/media/plotly/test_plot_20_b60ea359.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182138-gmmqn5n2/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: gmmqn5n2 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ltljjgyz with config: batch_size: 2 forecast_history: 15 lr: 0.001 number_encoder_layers: 3 out_seq_length: 5 use_mask: False wandb: Agent Started Run: ltljjgyz
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ltljjgyz INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmx0bGpqZ3l6OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/media/graph/graph_0_summary_e40e6689.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/media/graph
The running loss is: 19.090508729219437 The number of items in train is: 23 The loss for epoch 0 0.8300221186617146
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-history.jsonl
The running loss is: 18.449677228927612 The number of items in train is: 23 The loss for epoch 1 0.8021598795185918 The running loss is: 13.39623960852623 The number of items in train is: 23
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-summary.json
The loss for epoch 2 0.5824452003707057
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-history.jsonl
The running loss is: 13.160676665604115 The number of items in train is: 23 The loss for epoch 3 0.5722033332871355 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-history.jsonl
The running loss is: 12.167915090918541 The number of items in train is: 23 The loss for epoch 4 0.5290397865616757
INFO:wandb.wandb_agent:Running runs: ['ltljjgyz'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-history.jsonl
The running loss is: 8.957148090004921 The number of items in train is: 23 The loss for epoch 5 0.38944122130456177 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-history.jsonl
The running loss is: 6.491320692002773 The number of items in train is: 23 The loss for epoch 6 0.28223133443490317 2 The running loss is: 6.585609849542379 The number of items in train is: 23 The loss for epoch 7 0.28633086302358174
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-history.jsonl
The running loss is: 20.37033998966217 The number of items in train is: 23 The loss for epoch 8 0.8856669560722683
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-history.jsonl
The running loss is: 13.08682893961668 The number of items in train is: 23 The loss for epoch 9 0.5689925625920296 Data saved to: 28_May_202006_22PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_22PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 439.419891
66 2020-04-22 sub_region ... 66 439.420502
67 2020-04-23 sub_region ... 67 439.421204
68 2020-04-24 sub_region ... 68 439.428345
69 2020-04-25 sub_region ... 69 439.419373
70 2020-04-26 sub_region ... 70 439.426117
71 2020-04-27 sub_region ... 71 439.426697
72 2020-04-28 sub_region ... 72 439.417877
73 2020-04-29 sub_region ... 73 439.417023
74 2020-04-30 sub_region ... 74 439.419281
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/media/plotly/test_plot_20_2f23849f.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ltljjgyz
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182155-ltljjgyz/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ltljjgyz INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: hh6xz8uc with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 1 use_mask: True wandb: Agent Started Run: hh6xz8uc
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/hh6xz8uc INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmhoNnh6OHVjOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
13.29942782886792
The number of items in train is:
25
The loss for epoch 0
0.5319771131547167
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/media/graph/graph_0_summary_f49bd55b.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/media
The running loss is: 15.958505789749324 The number of items in train is: 25 The loss for epoch 1 0.638340231589973 The running loss is: 9.610189565457404 The number of items in train is: 25 The loss for epoch 2 0.38440758261829616 The running loss is: 8.331762389920186 The number of items in train is: 25
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-history.jsonl
The loss for epoch 3 0.33327049559680744 1 The running loss is: 6.60916328427993 The number of items in train is: 25 The loss for epoch 4 0.2643665313711972 2 The running loss is: 6.482465127395699 The number of items in train is: 25 The loss for epoch 5 0.25929860509582797 The running loss is: 7.1515498894732445 The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-summary.json
25
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-history.jsonl
The loss for epoch 6 0.2860619955789298 The running loss is: 6.881212463777047 The number of items in train is: 25 The loss for epoch 7 0.27524849855108185 The running loss is: 7.631635777652264 The number of items in train is: 25 The loss for epoch 8 0.30526543110609056 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-history.jsonl
The running loss is: 6.2557811347651295 The number of items in train is: 25 The loss for epoch 9 0.2502312453906052 Data saved to: 28_May_202006_22PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_22PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['hh6xz8uc'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 801.962341
66 2020-04-22 sub_region ... 66 824.838013
67 2020-04-23 sub_region ... 67 775.098450
68 2020-04-24 sub_region ... 68 820.728394
69 2020-04-25 sub_region ... 69 858.230591
70 2020-04-26 sub_region ... 70 790.655151
71 2020-04-27 sub_region ... 71 957.766296
72 2020-04-28 sub_region ... 72 793.469360
73 2020-04-29 sub_region ... 73 769.994263
74 2020-04-30 sub_region ... 74 771.788574
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/media/plotly/test_plot_20_302bf663.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: hh6xz8uc
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182215-hh6xz8uc/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: hh6xz8uc INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: f5uowdv7 with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 1 use_mask: False wandb: Agent Started Run: f5uowdv7
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/f5uowdv7 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmY1dW93ZHY3OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
13.510150267975405
The number of items in train is:
25
The loss for epoch 0
0.5404060107190162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/media/graph/graph_0_summary_146b15c3.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/media
The running loss is: 16.415556019172072 The number of items in train is: 25 The loss for epoch 1 0.6566222407668829 1 The running loss is: 8.791841979138553 The number of items in train is: 25 The loss for epoch 2 0.35167367916554215
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-history.jsonl
The running loss is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-summary.json
9.764055475010537 The number of items in train is: 25 The loss for epoch 3 0.3905622190004215 The running loss is: 7.065425713779405 The number of items in train is: 25 The loss for epoch 4 0.2826170285511762 1 The running loss is: 6.248815768616623 The number of items in train is: 25 The loss for epoch 5 0.24995263074466492
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-summary.json
The running loss is: 6.674299073871225 The number of items in train is: 25 The loss for epoch 6 0.266971962954849 1 The running loss is: 7.030096196685918 The number of items in train is: 25 The loss for epoch 7 0.2812038478674367 The running loss is: 8.11010273359716 The number of items in train is: 25 The loss for epoch 8 0.32440410934388636 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-summary.json
The running loss is: 8.296146783512086 The number of items in train is: 25 The loss for epoch 9 0.33184587134048343 Data saved to: 28_May_202006_22PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_22PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['f5uowdv7']
interpolate should be below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-summary.json
Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/config.yaml
CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 725.814941
66 2020-04-22 sub_region ... 66 739.527832
67 2020-04-23 sub_region ... 67 716.998169
68 2020-04-24 sub_region ... 68 738.462158
69 2020-04-25 sub_region ... 69 756.830688
70 2020-04-26 sub_region ... 70 728.941650
71 2020-04-27 sub_region ... 71 807.169800
72 2020-04-28 sub_region ... 72 725.725586
73 2020-04-29 sub_region ... 73 714.828308
74 2020-04-30 sub_region ... 74 716.417358
[25 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/media/plotly/test_plot_20_f6cc8bc8.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: f5uowdv7
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182233-f5uowdv7/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: f5uowdv7 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: o9ibe0l0 with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 2 use_mask: True wandb: Agent Started Run: o9ibe0l0
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/o9ibe0l0 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm85aWJlMGwwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
19.00246888399124
The number of items in train is:
25
The loss for epoch 0
0.7600987553596497
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/media/graph/graph_0_summary_5b5b7fc6.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/media/graph
The running loss is: 16.036757367663085 The number of items in train is: 25 The loss for epoch 1 0.6414702947065234 1 The running loss is: 10.140504654031247 The number of items in train is: 25 The loss for epoch 2 0.4056201861612499 The running loss is: 11.365801957435906 The number of items in train is: 25 The loss for epoch 3 0.45463207829743624
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-summary.json
1 The running loss is: 15.48135045915842 The number of items in train is: 25 The loss for epoch 4 0.6192540183663369 2 The running loss is: 8.226837230846286 The number of items in train is: 25 The loss for epoch 5 0.32907348923385144 The running loss is: 7.584667568095028 The number of items in train is: 25 The loss for epoch 6 0.30338670272380114
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-summary.json
1 The running loss is: 8.201896103098989 The number of items in train is: 25 The loss for epoch 7 0.3280758441239595 2 The running loss is: 8.102215321734548 The number of items in train is: 25 The loss for epoch 8 0.3240886128693819 The running loss is: 8.849566355347633 The number of items in train is: 25 The loss for epoch 9 0.35398265421390535
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-summary.json
1 Data saved to: 28_May_202006_22PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_22PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['o9ibe0l0']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9])
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-history.jsonl
Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-summary.json /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 721.418884
66 2020-04-22 sub_region ... 66 729.418823
67 2020-04-23 sub_region ... 67 704.071716
68 2020-04-24 sub_region ... 68 698.017456
69 2020-04-25 sub_region ... 69 721.563843
70 2020-04-26 sub_region ... 70 702.487244
71 2020-04-27 sub_region ... 71 729.263550
72 2020-04-28 sub_region ... 72 726.579834
73 2020-04-29 sub_region ... 73 722.222656
74 2020-04-30 sub_region ... 74 718.942139
[25 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/media/plotly/test_plot_20_359c7a7b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: o9ibe0l0
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182245-o9ibe0l0/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: o9ibe0l0 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: i4hyn3mi with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 2 use_mask: False wandb: Agent Started Run: i4hyn3mi
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/i4hyn3mi INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmk0aHluM21pOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
18.724818095564842
The number of items in train is:
25
The loss for epoch 0
0.7489927238225937
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/media/graph/graph_0_summary_a1e7384c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/media
The running loss is: 15.727116253226995 The number of items in train is: 25 The loss for epoch 1 0.6290846501290798 1 The running loss is: 8.56138514901977 The number of items in train is: 25 The loss for epoch 2 0.3424554059607908 The running loss is: 10.345093304291368 The number of items in train is: 25 The loss for epoch 3 0.4138037321716547
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-history.jsonl
1 The running loss is: 10.967490065842867 The number of items in train is: 25 The loss for epoch 4 0.43869960263371466 The running loss is: 7.611721908673644 The number of items in train is: 25 The loss for epoch 5 0.30446887634694575 The running loss is: 7.797869371250272 The number of items in train is: 25 The loss for epoch 6 0.31191477485001085
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-history.jsonl
The running loss is: 7.795289715752006 The number of items in train is: 25 The loss for epoch 7 0.31181158863008024 The running loss is: 6.245977671816945 The number of items in train is: 25 The loss for epoch 8 0.2498391068726778 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-history.jsonl
The running loss is: 6.67815606854856 The number of items in train is: 25 The loss for epoch 9 0.2671262427419424 Data saved to: 28_May_202006_23PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_23PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
INFO:wandb.wandb_agent:Running runs: ['i4hyn3mi']
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 648.033325
66 2020-04-22 sub_region ... 66 696.399780
67 2020-04-23 sub_region ... 67 633.036682
68 2020-04-24 sub_region ... 68 599.434021
69 2020-04-25 sub_region ... 69 679.843323
70 2020-04-26 sub_region ... 70 630.065979
71 2020-04-27 sub_region ... 71 673.871094
72 2020-04-28 sub_region ... 72 681.052856
73 2020-04-29 sub_region ... 73 670.992432
74 2020-04-30 sub_region ... 74 661.747437
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-history.jsonl
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: i4hyn3mi
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/media/plotly/test_plot_20_c2722b2a.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182258-i4hyn3mi/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: i4hyn3mi INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: lcu7qwcc with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 3 use_mask: True wandb: Agent Started Run: lcu7qwcc
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/lcu7qwcc INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmxjdTdxd2NjOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
12.417334482073784
The number of items in train is:
24
The loss for epoch 0
0.5173889367530743
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/media/graph/graph_0_summary_2e285db2.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/media/graph
The running loss is: 14.187367584556341 The number of items in train is: 24 The loss for epoch 1 0.5911403160231808 The running loss is: 8.399394869804382 The number of items in train is: 24 The loss for epoch 2 0.34997478624184925 1 The running loss is: 6.990775217302144 The number of items in train is: 24 The loss for epoch 3 0.29128230072092265
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-summary.json
The running loss is: 8.946204421576113 The number of items in train is: 24 The loss for epoch 4 0.37275851756567135 1 The running loss is: 15.735674642026424 The number of items in train is: 24 The loss for epoch 5 0.6556531100844344 The running loss is: 7.7431636694818735 The number of items in train is: 24 The loss for epoch 6 0.32263181956174475
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-summary.json
1 The running loss is: 4.328124134801328 The number of items in train is: 24 The loss for epoch 7 0.18033850561672202 2 The running loss is: 3.735455541405827 The number of items in train is: 24 The loss for epoch 8 0.15564398089190945
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-summary.json
The running loss is: 3.9662495320662856 The number of items in train is: 24 The loss for epoch 9 0.16526039716942856 1 Data saved to: 28_May_202006_23PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_23PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['lcu7qwcc']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 755.517212
66 2020-04-22 sub_region ... 66 771.916016
67 2020-04-23 sub_region ... 67 760.687378
68 2020-04-24 sub_region ... 68 735.305786
69 2020-04-25 sub_region ... 69 728.676880
70 2020-04-26 sub_region ... 70 742.574097
71 2020-04-27 sub_region ... 71 690.534424
72 2020-04-28 sub_region ... 72 759.479492
73 2020-04-29 sub_region ... 73 761.484863
74 2020-04-30 sub_region ... 74 765.487183
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/media/plotly/test_plot_20_1d2f2943.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: lcu7qwcc
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182310-lcu7qwcc/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: lcu7qwcc INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: cra6n8ca with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 3 use_mask: False wandb: Agent Started Run: cra6n8ca
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/cra6n8ca INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmNyYTZuOGNhOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
12.737247206270695
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/config.yaml
The number of items in train is: 24 The loss for epoch 0 0.5307186335946122
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/media/graph/graph_0_summary_1b024392.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/media/graph
The running loss is: 12.846482746303082 The number of items in train is: 24 The loss for epoch 1 0.5352701144292951 The running loss is: 8.131057146936655 The number of items in train is: 24 The loss for epoch 2 0.3387940477890273 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/wandb-summary.json
The running loss is: 5.895240487996489 The number of items in train is: 24 The loss for epoch 3 0.24563502033318704 2 The running loss is: 8.398755688220263 The number of items in train is: 24 The loss for epoch 4 0.34994815367584425 The running loss is: 16.694155018776655 The number of items in train is: 24 The loss for epoch 5 0.6955897924490273
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/wandb-summary.json
The running loss is: 9.139933694154024 The number of items in train is: 24 The loss for epoch 6 0.380830570589751 1 The running loss is: 5.126268687658012 The number of items in train is: 24 The loss for epoch 7 0.21359452865241715 2 The running loss is: 4.945361414458603 The number of items in train is: 24 The loss for epoch 8 0.2060567256024418 3 Stopping model now Data saved to: 28_May_202006_23PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/wandb-summary.json
Data saved to: 28_May_202006_23PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['cra6n8ca']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 596.721497
66 2020-04-22 sub_region ... 66 611.140503
67 2020-04-23 sub_region ... 67 602.298889
68 2020-04-24 sub_region ... 68 592.272522
69 2020-04-25 sub_region ... 69 590.910889
70 2020-04-26 sub_region ... 70 592.348145
71 2020-04-27 sub_region ... 71 565.893555
72 2020-04-28 sub_region ... 72 597.307617
73 2020-04-29 sub_region ... 73 598.141846
74 2020-04-30 sub_region ... 74 604.241028
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: cra6n8ca
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/media/plotly/test_plot_all_19_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/media/plotly/test_plot_18_b0d51432.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182325-cra6n8ca/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: cra6n8ca INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: kbi41ndm with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 4 use_mask: True wandb: Agent Started Run: kbi41ndm
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/kbi41ndm INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmtiaTQxbmRtOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
15.76520562171936
The number of items in train is:
24
The loss for epoch 0
0.65688356757164
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/media/graph/graph_0_summary_e02759de.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/media/graph
The running loss is: 15.477864772081375 The number of items in train is: 24 The loss for epoch 1 0.6449110321700573 1 The running loss is: 12.414291091263294 The number of items in train is: 24 The loss for epoch 2 0.5172621288026372 2 The running loss is: 7.5504014934413135 The number of items in train is: 24 The loss for epoch 3 0.3146000622267214
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-summary.json
The running loss is: 12.823520310223103 The number of items in train is: 24 The loss for epoch 4 0.5343133462592959 1 The running loss is: 6.759511549025774 The number of items in train is: 24 The loss for epoch 5 0.2816463145427406 The running loss is: 6.30719225294888 The number of items in train is: 24 The loss for epoch 6 0.26279967720620334 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-summary.json
The running loss is: 6.85306416079402 The number of items in train is: 24 The loss for epoch 7 0.28554434003308415 2 The running loss is: 6.9310750886797905 The number of items in train is: 24 The loss for epoch 8 0.2887947953616579 The running loss is: 8.455319091677666 The number of items in train is: 24 The loss for epoch 9 0.3523049621532361 1 Data saved to: 28_May_202006_23PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-summary.json
Data saved to: 28_May_202006_23PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['kbi41ndm']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 779.131897
66 2020-04-22 sub_region ... 66 792.555298
67 2020-04-23 sub_region ... 67 780.039795
68 2020-04-24 sub_region ... 68 769.559204
69 2020-04-25 sub_region ... 69 768.763855
70 2020-04-26 sub_region ... 70 775.261108
71 2020-04-27 sub_region ... 71 753.575806
72 2020-04-28 sub_region ... 72 784.202087
73 2020-04-29 sub_region ... 73 783.728699
74 2020-04-30 sub_region ... 74 785.233765
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/media/plotly/test_plot_20_f7c7c22f.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/media/plotly
wandb: Agent Finished Run: kbi41ndm
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182343-kbi41ndm/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: kbi41ndm INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: tcdi2b90 with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 4 use_mask: False wandb: Agent Started Run: tcdi2b90
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/tcdi2b90 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnRjZGkyYjkwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
15.728306412696838
The number of items in train is:
24
The loss for epoch 0
0.655346100529035
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/media/graph/graph_0_summary_2e266998.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/media
The running loss is: 13.534287229180336 The number of items in train is: 24 The loss for epoch 1 0.5639286345491806 1 The running loss is: 10.27218734100461 The number of items in train is: 24 The loss for epoch 2 0.4280078058751921 2 The running loss is: 5.5147014474496245 The number of items in train is: 24 The loss for epoch 3 0.22977922697706768
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-history.jsonl
The running loss is: 9.376355066895485 The number of items in train is: 24 The loss for epoch 4 0.3906814611206452 1 The running loss is: 6.485733915120363 The number of items in train is: 24 The loss for epoch 5 0.27023891313001513 2 The running loss is: 7.979923363775015 The number of items in train is: 24 The loss for epoch 6 0.33249680682395893
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-history.jsonl
The running loss is: 8.143040359020233 The number of items in train is: 24 The loss for epoch 7 0.33929334829250973 1 The running loss is: 5.536451553925872 The number of items in train is: 24 The loss for epoch 8 0.230685481413578 2 The running loss is: 4.550636069383472 The number of items in train is: 24 The loss for epoch 9 0.18960983622431135 Data saved to: 28_May_202006_23PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-history.jsonl
Data saved to: 28_May_202006_23PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['tcdi2b90']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 756.924194
66 2020-04-22 sub_region ... 66 788.874878
67 2020-04-23 sub_region ... 67 767.624146
68 2020-04-24 sub_region ... 68 745.193481
69 2020-04-25 sub_region ... 69 751.710815
70 2020-04-26 sub_region ... 70 755.871094
71 2020-04-27 sub_region ... 71 710.098877
72 2020-04-28 sub_region ... 72 772.029785
73 2020-04-29 sub_region ... 73 776.352173
74 2020-04-30 sub_region ... 74 783.972778
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/media/plotly/test_plot_20_75431aa1.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: tcdi2b90
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182355-tcdi2b90/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: tcdi2b90 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 9p7a2kaq with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 5 use_mask: True wandb: Agent Started Run: 9p7a2kaq
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/9p7a2kaq INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjlwN2Eya2FxOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
12.287862300872803
The number of items in train is:
23
The loss for epoch 0
0.5342548826466436
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/media/graph/graph_0_summary_220843b4.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/media
The running loss is: 13.926748000085354 The number of items in train is: 23 The loss for epoch 1 0.6055107826124067 1 The running loss is: 11.354672580957413 The number of items in train is: 23 The loss for epoch 2 0.4936814165633658 2 The running loss is: 9.702557804062963 The number of items in train is: 23 The loss for epoch 3 0.4218503393070853
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/wandb-summary.json
The running loss is: 11.627340912818909 The number of items in train is: 23 The loss for epoch 4 0.5055365614269091 1 The running loss is: 6.340536650270224 The number of items in train is: 23 The loss for epoch 5 0.275675506533488 2 The running loss is: 10.078678345307708 The number of items in train is: 23 The loss for epoch 6 0.4382034063177264
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/wandb-summary.json
The running loss is: 8.461791813373566 The number of items in train is: 23 The loss for epoch 7 0.3679039918858072 1 The running loss is: 4.9823848977684975 The number of items in train is: 23 The loss for epoch 8 0.21662543033776077 2 The running loss is: 3.317102179862559 The number of items in train is: 23 The loss for epoch 9 0.14422183390706778 Data saved to: 28_May_202006_24PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/wandb-summary.json
Data saved to: 28_May_202006_24PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 835.124390
66 2020-04-22 sub_region ... 66 840.772278
67 2020-04-23 sub_region ... 67 832.618164
68 2020-04-24 sub_region ... 68 835.653015
69 2020-04-25 sub_region ... 69 821.924072
70 2020-04-26 sub_region ... 70 833.929749
71 2020-04-27 sub_region ... 71 823.226929
72 2020-04-28 sub_region ... 72 831.859680
73 2020-04-29 sub_region ... 73 828.141541
74 2020-04-30 sub_region ... 74 831.716797
[25 rows x 28 columns]
INFO:wandb.wandb_agent:Running runs: ['9p7a2kaq']
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 9p7a2kaq
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/media/plotly/test_plot_20_7a7a1d5c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182412-9p7a2kaq/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 9p7a2kaq INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: a7xv42ra with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 1 out_seq_length: 5 use_mask: False wandb: Agent Started Run: a7xv42ra
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/a7xv42ra INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmE3eHY0MnJhOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
12.292471200227737
The number of items in train is:
23
The loss for epoch 0
0.534455269575119
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/media/graph/graph_0_summary_adccc66b.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/media
The running loss is: 14.315681867301464 The number of items in train is: 23 The loss for epoch 1 0.6224209507522376 1 The running loss is: 10.289977289736271 The number of items in train is: 23 The loss for epoch 2 0.44739031694505527 2 The running loss is: 7.302962388843298 The number of items in train is: 23 The loss for epoch 3 0.3175201038627521
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/wandb-summary.json
The running loss is: 10.639127224683762 The number of items in train is: 23 The loss for epoch 4 0.462570748899294 The running loss is: 5.083698073402047 The number of items in train is: 23 The loss for epoch 5 0.2210303510174803 1 The running loss is: 4.266114007681608 The number of items in train is: 23 The loss for epoch 6 0.1854832177252873 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/wandb-summary.json
The running loss is: 3.362240646034479 The number of items in train is: 23 The loss for epoch 7 0.14618437591454256 The running loss is: 5.346718622371554 The number of items in train is: 23 The loss for epoch 8 0.2324660270596328 1 The running loss is: 3.976900838315487 The number of items in train is: 23 The loss for epoch 9 0.17290873210067334 Data saved to: 28_May_202006_24PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/wandb-summary.json
Data saved to: 28_May_202006_24PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() INFO:wandb.wandb_agent:Running runs: ['a7xv42ra'] /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 773.226318
66 2020-04-22 sub_region ... 66 785.933838
67 2020-04-23 sub_region ... 67 776.867065
68 2020-04-24 sub_region ... 68 779.012268
69 2020-04-25 sub_region ... 69 770.271240
70 2020-04-26 sub_region ... 70 779.572510
71 2020-04-27 sub_region ... 71 765.994873
72 2020-04-28 sub_region ... 72 776.860962
73 2020-04-29 sub_region ... 73 774.669739
74 2020-04-30 sub_region ... 74 777.345093
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: a7xv42ra
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/media/plotly/test_plot_20_b982c2b4.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182423-a7xv42ra/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: a7xv42ra INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 0c0zfqxx with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 1 use_mask: True wandb: Agent Started Run: 0c0zfqxx
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/0c0zfqxx INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjBjMHpmcXh4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/media/graph/graph_0_summary_4b5baab2.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/media
The running loss is: 10.145409000921063 The number of items in train is: 25 The loss for epoch 0 0.4058163600368425 The running loss is: 25.701497849076986 The number of items in train is: 25 The loss for epoch 1 1.0280599139630795
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-summary.json
The running loss is: 20.272816001903266 The number of items in train is: 25 The loss for epoch 2 0.8109126400761306
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-summary.json
The running loss is: 19.751347083598375 The number of items in train is: 25 The loss for epoch 3 0.790053883343935 The running loss is: 18.793059289455414 The number of items in train is: 25 The loss for epoch 4 0.7517223715782165 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-summary.json
The running loss is: 16.612464647740126 The number of items in train is: 25 The loss for epoch 5 0.664498585909605 2
INFO:wandb.wandb_agent:Running runs: ['0c0zfqxx']
The running loss is: 9.692850414896384 The number of items in train is: 25 The loss for epoch 6 0.38771401659585536
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-summary.json
The running loss is: 19.73813858255744 The number of items in train is: 25 The loss for epoch 7 0.7895255433022976
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-summary.json
The running loss is: 12.20803095260635 The number of items in train is: 25 The loss for epoch 8 0.488321238104254 1 The running loss is: 11.476102620712481 The number of items in train is: 25 The loss for epoch 9 0.4590441048284993 2 Data saved to: 28_May_202006_24PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-summary.json
Data saved to: 28_May_202006_24PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 801.515869
66 2020-04-22 sub_region ... 66 801.592773
67 2020-04-23 sub_region ... 67 801.521545
68 2020-04-24 sub_region ... 68 801.597168
69 2020-04-25 sub_region ... 69 801.738037
70 2020-04-26 sub_region ... 70 801.605835
71 2020-04-27 sub_region ... 71 801.837097
72 2020-04-28 sub_region ... 72 801.528931
73 2020-04-29 sub_region ... 73 801.504639
74 2020-04-30 sub_region ... 74 801.555054
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 0c0zfqxx
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/media/plotly/test_plot_20_fcc89d57.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182436-0c0zfqxx/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 0c0zfqxx INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ozprnvr9 with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 1 use_mask: False wandb: Agent Started Run: ozprnvr9
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ozprnvr9 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm96cHJudnI5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/media/graph/graph_0_summary_9dca8cf5.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/media/graph
The running loss is: 10.096708084311103 The number of items in train is: 25 The loss for epoch 0 0.40386832337244416 The running loss is: 25.173049704171717 The number of items in train is: 25 The loss for epoch 1 1.0069219881668687
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-summary.json
The running loss is: 20.567884587217122 The number of items in train is: 25 The loss for epoch 2 0.8227153834886849
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-summary.json
The running loss is: 16.932578061707318 The number of items in train is: 25 The loss for epoch 3 0.6773031224682927 1 The running loss is: 11.519542964175344 The number of items in train is: 25 The loss for epoch 4 0.4607817185670137 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-summary.json
The running loss is: 8.808077305089682 The number of items in train is: 25 The loss for epoch 5 0.3523230922035873
INFO:wandb.wandb_agent:Running runs: ['ozprnvr9']
The running loss is: 11.38830050965771 The number of items in train is: 25 The loss for epoch 6 0.4555320203863084 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-summary.json
The running loss is: 7.139500390738249 The number of items in train is: 25 The loss for epoch 7 0.28558001562952995 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-summary.json
The running loss is: 9.381748400162905 The number of items in train is: 25 The loss for epoch 8 0.3752699360065162 The running loss is: 10.521757567301393 The number of items in train is: 25 The loss for epoch 9 0.4208703026920557 Data saved to: 28_May_202006_25PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-summary.json
Data saved to: 28_May_202006_25PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 761.296265
66 2020-04-22 sub_region ... 66 761.453735
67 2020-04-23 sub_region ... 67 761.306885
68 2020-04-24 sub_region ... 68 761.469482
69 2020-04-25 sub_region ... 69 761.735107
70 2020-04-26 sub_region ... 70 761.236694
71 2020-04-27 sub_region ... 71 761.615662
72 2020-04-28 sub_region ... 72 760.781616
73 2020-04-29 sub_region ... 73 760.612366
74 2020-04-30 sub_region ... 74 760.868896
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/media/plotly/test_plot_20_b3c2d6a3.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ozprnvr9
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182453-ozprnvr9/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ozprnvr9 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: wgc5r97k with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 2 use_mask: True wandb: Agent Started Run: wgc5r97k
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/wgc5r97k INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOndnYzVyOTdrOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/media/graph/graph_0_summary_f21185d8.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/media/graph
The running loss is: 20.584014743566513 The number of items in train is: 25 The loss for epoch 0 0.8233605897426606 The running loss is: 17.932957649230957 The number of items in train is: 25 The loss for epoch 1 0.7173183059692383 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-history.jsonl
The running loss is: 15.548911690013483 The number of items in train is: 25 The loss for epoch 2 0.6219564676005394
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-history.jsonl
The running loss is: 14.730191620066762 The number of items in train is: 25 The loss for epoch 3 0.5892076648026705 The running loss is: 15.559959415346384 The number of items in train is: 25 The loss for epoch 4 0.6223983766138553 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-history.jsonl
The running loss is: 11.396017402410507 The number of items in train is: 25 The loss for epoch 5 0.4558406960964203 2
INFO:wandb.wandb_agent:Running runs: ['wgc5r97k']
The running loss is: 9.876246387138963 The number of items in train is: 25 The loss for epoch 6 0.3950498554855585 3 Stopping model now Data saved to: 28_May_202006_25PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-history.jsonl
Data saved to: 28_May_202006_25PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 505.433014
66 2020-04-22 sub_region ... 66 504.062500
67 2020-04-23 sub_region ... 67 504.517761
68 2020-04-24 sub_region ... 68 503.535156
69 2020-04-25 sub_region ... 69 506.940369
70 2020-04-26 sub_region ... 70 505.692200
71 2020-04-27 sub_region ... 71 506.551117
72 2020-04-28 sub_region ... 72 507.229309
73 2020-04-29 sub_region ... 73 507.427246
74 2020-04-30 sub_region ... 74 507.579987
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/media/plotly/test_plot_14_c7166984.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/media/plotly
wandb: Agent Finished Run: wgc5r97k
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/media/plotly/test_plot_all_15_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182514-wgc5r97k/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: wgc5r97k INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 5c9wn3n7 with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 2 use_mask: False wandb: Agent Started Run: 5c9wn3n7
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/5c9wn3n7 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjVjOXduM243OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/media/graph/graph_0_summary_698fbc23.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/media
The running loss is: 20.51875665783882 The number of items in train is: 25 The loss for epoch 0 0.8207502663135529 The running loss is: 18.745911203324795 The number of items in train is: 25 The loss for epoch 1 0.7498364481329918 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-summary.json
The running loss is: 14.171599298715591 The number of items in train is: 25 The loss for epoch 2 0.5668639719486237 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-summary.json
The running loss is: 10.58555408520624 The number of items in train is: 25 The loss for epoch 3 0.4234221634082496 The running loss is: 17.37247646227479 The number of items in train is: 25 The loss for epoch 4 0.6948990584909915
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-summary.json
The running loss is: 10.633864669129252 The number of items in train is: 25 The loss for epoch 5 0.4253545867651701 1
INFO:wandb.wandb_agent:Running runs: ['5c9wn3n7']
The running loss is: 9.219570185989141 The number of items in train is: 25 The loss for epoch 6 0.36878280743956565 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-summary.json
The running loss is: 8.460247997194529 The number of items in train is: 25 The loss for epoch 7 0.33840991988778113 3 Stopping model now Data saved to: 28_May_202006_25PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_25PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 684.219727
66 2020-04-22 sub_region ... 66 684.323364
67 2020-04-23 sub_region ... 67 684.080750
68 2020-04-24 sub_region ... 68 683.885254
69 2020-04-25 sub_region ... 69 685.422363
70 2020-04-26 sub_region ... 70 684.204346
71 2020-04-27 sub_region ... 71 685.634094
72 2020-04-28 sub_region ... 72 684.494141
73 2020-04-29 sub_region ... 73 684.385315
74 2020-04-30 sub_region ... 74 684.536133
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/media/plotly/test_plot_16_e1926f30.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 5c9wn3n7
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/media/plotly/test_plot_all_17_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182533-5c9wn3n7/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 5c9wn3n7 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 63viw4xu with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 3 use_mask: True wandb: Agent Started Run: 63viw4xu
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/63viw4xu INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjYzdml3NHh1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/media/graph/graph_0_summary_3f6952c2.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/media
The running loss is: 18.772414907813072 The number of items in train is: 24 The loss for epoch 0 0.7821839544922113 The running loss is: 19.575059436261654 The number of items in train is: 24 The loss for epoch 1 0.8156274765109023
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-history.jsonl
The running loss is: 15.316046819090843 The number of items in train is: 24 The loss for epoch 2 0.6381686174621185 The running loss is: 11.947806634008884 The number of items in train is: 24 The loss for epoch 3 0.49782527641703683
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-history.jsonl
The running loss is: 10.35117731243372 The number of items in train is: 24 The loss for epoch 4 0.43129905468473834 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-history.jsonl
The running loss is: 11.100783228874207 The number of items in train is: 24 The loss for epoch 5 0.4625326345364253 2 The running loss is: 11.392054236494005 The number of items in train is: 24 The loss for epoch 6 0.4746689265205835
INFO:wandb.wandb_agent:Running runs: ['63viw4xu'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-history.jsonl
The running loss is: 5.173449165187776 The number of items in train is: 24 The loss for epoch 7 0.215560381882824 1 The running loss is: 9.032942300196737 The number of items in train is: 24 The loss for epoch 8 0.3763725958415307
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-history.jsonl
The running loss is: 19.382783502340317 The number of items in train is: 24 The loss for epoch 9 0.8076159792641798 Data saved to: 28_May_202006_25PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_25PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 515.919434
66 2020-04-22 sub_region ... 66 516.543030
67 2020-04-23 sub_region ... 67 515.078186
68 2020-04-24 sub_region ... 68 516.592896
69 2020-04-25 sub_region ... 69 515.011414
70 2020-04-26 sub_region ... 70 514.803894
71 2020-04-27 sub_region ... 71 516.728943
72 2020-04-28 sub_region ... 72 515.620483
73 2020-04-29 sub_region ... 73 515.851013
74 2020-04-30 sub_region ... 74 514.158508
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/media/plotly/test_plot_20_14ad17d9.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 63viw4xu
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182550-63viw4xu/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 63viw4xu INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: qhdtbeld with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 3 use_mask: False wandb: Agent Started Run: qhdtbeld
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/qhdtbeld INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnFoZHRiZWxkOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/media/graph/graph_0_summary_14707cc8.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/media/graph
The running loss is: 18.797405660152435 The number of items in train is: 24 The loss for epoch 0 0.7832252358396848 The running loss is: 19.402401946485043 The number of items in train is: 24 The loss for epoch 1 0.8084334144368768
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-history.jsonl
The running loss is: 14.805835217237473 The number of items in train is: 24 The loss for epoch 2 0.616909800718228 The running loss is: 9.891540326178074 The number of items in train is: 24 The loss for epoch 3 0.4121475135907531 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-history.jsonl
The running loss is: 5.6078918017446995 The number of items in train is: 24 The loss for epoch 4 0.23366215840602914 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-history.jsonl
The running loss is: 4.041457449435256 The number of items in train is: 24 The loss for epoch 5 0.16839406039313567 3 Stopping model now Data saved to: 28_May_202006_26PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_26PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['qhdtbeld']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 378.001099
66 2020-04-22 sub_region ... 66 379.296204
67 2020-04-23 sub_region ... 67 378.866272
68 2020-04-24 sub_region ... 68 383.898071
69 2020-04-25 sub_region ... 69 369.884155
70 2020-04-26 sub_region ... 70 381.321381
71 2020-04-27 sub_region ... 71 371.926025
72 2020-04-28 sub_region ... 72 378.183258
73 2020-04-29 sub_region ... 73 379.500549
74 2020-04-30 sub_region ... 74 378.237946
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/media/plotly/test_plot_12_9498e4f6.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: qhdtbeld
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/media/plotly/test_plot_all_13_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182608-qhdtbeld/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: qhdtbeld INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: mpoj2ncv with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 4 use_mask: True wandb: Agent Started Run: mpoj2ncv
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/mpoj2ncv INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm1wb2oybmN2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/media/graph/graph_0_summary_3a375e0b.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/media/graph
The running loss is: 21.451373293995857 The number of items in train is: 24 The loss for epoch 0 0.8938072205831608 The running loss is: 18.959119454026222 The number of items in train is: 24 The loss for epoch 1 0.7899633105844259
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-history.jsonl
The running loss is: 16.811535373330116 The number of items in train is: 24 The loss for epoch 2 0.7004806405554215 The running loss is: 15.114799819886684 The number of items in train is: 24 The loss for epoch 3 0.6297833258286119 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-history.jsonl
The running loss is: 13.837378360331059 The number of items in train is: 24 The loss for epoch 4 0.5765574316804608 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-history.jsonl
The running loss is: 10.938715167343616 The number of items in train is: 24 The loss for epoch 5 0.45577979863931734 3 Stopping model now Data saved to: 28_May_202006_26PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_26PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['mpoj2ncv'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 379.669922
66 2020-04-22 sub_region ... 66 381.567261
67 2020-04-23 sub_region ... 67 381.375244
68 2020-04-24 sub_region ... 68 386.184875
69 2020-04-25 sub_region ... 69 368.961060
70 2020-04-26 sub_region ... 70 383.574829
71 2020-04-27 sub_region ... 71 370.266907
72 2020-04-28 sub_region ... 72 379.182159
73 2020-04-29 sub_region ... 73 380.702667
74 2020-04-30 sub_region ... 74 380.433716
[25 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/media/plotly/test_plot_12_ab9062b5.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: mpoj2ncv
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/media/plotly/test_plot_all_13_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182620-mpoj2ncv/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: mpoj2ncv INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: em108twu with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 4 use_mask: False wandb: Agent Started Run: em108twu
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/em108twu INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmVtMTA4dHd1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/media/graph/graph_0_summary_46e50bbe.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/media/graph
The running loss is: 21.45643301308155 The number of items in train is: 24 The loss for epoch 0 0.8940180422117313 The running loss is: 18.878746390342712 The number of items in train is: 24 The loss for epoch 1 0.7866144329309464
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-history.jsonl
The running loss is: 16.909177042543888 The number of items in train is: 24 The loss for epoch 2 0.7045490434393287 The running loss is: 14.992017075419426 The number of items in train is: 24 The loss for epoch 3 0.6246673781424761 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-history.jsonl
The running loss is: 13.057133480906487 The number of items in train is: 24 The loss for epoch 4 0.5440472283711036 2 The running loss is: 7.168162252753973 The number of items in train is: 24 The loss for epoch 5
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-history.jsonl
0.2986734271980822 3 Stopping model now Data saved to: 28_May_202006_26PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_26PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['em108twu'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 381.720642
66 2020-04-22 sub_region ... 66 383.487915
67 2020-04-23 sub_region ... 67 382.618774
68 2020-04-24 sub_region ... 68 386.354279
69 2020-04-25 sub_region ... 69 371.824524
70 2020-04-26 sub_region ... 70 384.303436
71 2020-04-27 sub_region ... 71 373.255554
72 2020-04-28 sub_region ... 72 381.710022
73 2020-04-29 sub_region ... 73 383.322937
74 2020-04-30 sub_region ... 74 382.421906
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/media/plotly/test_plot_12_d4e101e3.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: em108twu
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/media/plotly/test_plot_all_13_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182632-em108twu/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: em108twu INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ecutz549 with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 5 use_mask: True wandb: Agent Started Run: ecutz549
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ecutz549 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmVjdXR6NTQ5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/media/graph/graph_0_summary_a3da4031.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/media
The running loss is: 17.729351297020912 The number of items in train is: 23 The loss for epoch 0 0.7708413607400396 The running loss is: 17.55451713502407 The number of items in train is: 23 The loss for epoch 1 0.7632398754358292
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-summary.json
The running loss is: 13.769226357340813 The number of items in train is: 23 The loss for epoch 2 0.5986620155365571 1 The running loss is: 10.689910009503365 The number of items in train is: 23 The loss for epoch 3 0.46477869606536365 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-summary.json
The running loss is: 10.719441205263138 The number of items in train is: 23 The loss for epoch 4 0.4660626610983973 The running loss is: 11.102143481373787 The number of items in train is: 23 The loss for epoch 5 0.48270189049451245 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-summary.json
The running loss is: 8.552160829305649 The number of items in train is: 23 The loss for epoch 6 0.3718330795350282
INFO:wandb.wandb_agent:Running runs: ['ecutz549']
The running loss is: 8.911982571706176 The number of items in train is: 23 The loss for epoch 7 0.3874775031176598
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-summary.json
1 The running loss is: 10.839222326874733 The number of items in train is: 23 The loss for epoch 8 0.4712705359510753 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-summary.json
The running loss is: 6.471042025834322 The number of items in train is: 23 The loss for epoch 9 0.28134965329714445 Data saved to: 28_May_202006_26PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_26PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 682.042175
66 2020-04-22 sub_region ... 66 684.554626
67 2020-04-23 sub_region ... 67 676.001160
68 2020-04-24 sub_region ... 68 688.336792
69 2020-04-25 sub_region ... 69 666.439087
70 2020-04-26 sub_region ... 70 671.417664
71 2020-04-27 sub_region ... 71 689.934692
72 2020-04-28 sub_region ... 72 674.941956
73 2020-04-29 sub_region ... 73 665.878967
74 2020-04-30 sub_region ... 74 658.025818
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/media/plotly/test_plot_20_790f76c3.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ecutz549
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182649-ecutz549/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ecutz549 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: m7ynw4z3 with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 2 out_seq_length: 5 use_mask: False wandb: Agent Started Run: m7ynw4z3
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/m7ynw4z3 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm03eW53NHozOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/config.yaml
The running loss is: 17.749317228794098 The number of items in train is:
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-metadata.json
23 The loss for epoch 0
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-events.jsonl
0.7717094447301782
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/media/graph/graph_0_summary_654b7d89.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/media
The running loss is: 17.551174700260162 The number of items in train is: 23 The loss for epoch 1 0.7630945521852245
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-history.jsonl
The running loss is: 13.54475101083517 The number of items in train is: 23 The loss for epoch 2 0.5889022178623987 1 The running loss is: 10.344177260994911 The number of items in train is: 23 The loss for epoch 3 0.44974683743456134 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-history.jsonl
The running loss is: 9.028607338666916 The number of items in train is: 23 The loss for epoch 4 0.39254814515943115 3 Stopping model now Data saved to: 28_May_202006_27PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_27PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 391.728210
66 2020-04-22 sub_region ... 66 392.369507
67 2020-04-23 sub_region ... 67 392.830872
68 2020-04-24 sub_region ... 68 397.064850
69 2020-04-25 sub_region ... 69 385.320221
70 2020-04-26 sub_region ... 70 393.841858
71 2020-04-27 sub_region ... 71 388.418610
72 2020-04-28 sub_region ... 72 390.187714
73 2020-04-29 sub_region ... 73 390.757416
74 2020-04-30 sub_region ... 74 391.191467
[25 rows x 28 columns]
INFO:wandb.wandb_agent:Running runs: ['m7ynw4z3']
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/media/plotly/test_plot_10_ca77e0e0.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: m7ynw4z3
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/media/plotly/test_plot_all_11_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182708-m7ynw4z3/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: m7ynw4z3 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ixe38oga with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 1 use_mask: True wandb: Agent Started Run: ixe38oga
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ixe38oga INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOml4ZTM4b2dhOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/media/graph/graph_0_summary_63db277c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/media/graph
The running loss is: 10.765136603731662 The number of items in train is: 25 The loss for epoch 0 0.4306054641492665
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-summary.json
The running loss is: 26.137935783714056 The number of items in train is: 25 The loss for epoch 1 1.0455174313485622
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-summary.json
The running loss is: 22.011536345118657 The number of items in train is: 25 The loss for epoch 2 0.8804614538047463
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-summary.json
The running loss is: 18.903898189775646 The number of items in train is: 25 The loss for epoch 3 0.7561559275910258
INFO:wandb.wandb_agent:Running runs: ['ixe38oga']
The running loss is: 18.865029871463776 The number of items in train is: 25 The loss for epoch 4 0.754601194858551
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-summary.json
The running loss is: 19.70023591836798 The number of items in train is: 25 The loss for epoch 5 0.7880094367347192 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-summary.json
The running loss is: 18.773742999881506 The number of items in train is: 25 The loss for epoch 6 0.7509497199952603
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-summary.json
The running loss is: 19.47792062163353 The number of items in train is: 25 The loss for epoch 7 0.7791168248653412
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-summary.json
The running loss is: 19.640535548329353 The number of items in train is: 25 The loss for epoch 8 0.7856214219331741 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-summary.json
The running loss is: 19.046687852591276 The number of items in train is: 25 The loss for epoch 9 0.7618675141036511 Data saved to: 28_May_202006_27PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_27PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 496.016388
66 2020-04-22 sub_region ... 66 496.046783
67 2020-04-23 sub_region ... 67 495.631775
68 2020-04-24 sub_region ... 68 495.853607
69 2020-04-25 sub_region ... 69 496.806030
70 2020-04-26 sub_region ... 70 495.802399
71 2020-04-27 sub_region ... 71 497.801544
72 2020-04-28 sub_region ... 72 496.134644
73 2020-04-29 sub_region ... 73 495.858582
74 2020-04-30 sub_region ... 74 495.841339
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: ixe38oga
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/media/plotly/test_plot_20_59b2352c.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182722-ixe38oga/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: ixe38oga INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 2g463y10 with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 1 use_mask: False wandb: Agent Started Run: 2g463y10
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/2g463y10 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjJnNDYzeTEwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/media/graph/graph_0_summary_42cee6f6.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/media
The running loss is: 11.082796600530855 The number of items in train is: 25 The loss for epoch 0 0.44331186402123424
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-summary.json
The running loss is: 26.767839308828115 The number of items in train is: 25 The loss for epoch 1 1.0707135723531247
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-summary.json
The running loss is: 19.91911711357534 The number of items in train is: 25 The loss for epoch 2 0.7967646845430135
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-summary.json
The running loss is: 18.481270607560873 The number of items in train is: 25 The loss for epoch 3 0.7392508243024349 1
INFO:wandb.wandb_agent:Running runs: ['2g463y10']
The running loss is: 18.93701122701168 The number of items in train is: 25 The loss for epoch 4 0.7574804490804672
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-summary.json
The running loss is: 19.590259796008468 The number of items in train is: 25 The loss for epoch 5 0.7836103918403388
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-summary.json
The running loss is: 18.897615873254836 The number of items in train is: 25 The loss for epoch 6 0.7559046349301934 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-summary.json
The running loss is: 19.6400615721941 The number of items in train is: 25 The loss for epoch 7 0.785602462887764
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-summary.json
The running loss is: 19.225644499063492 The number of items in train is: 25 The loss for epoch 8 0.7690257799625396 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-summary.json
The running loss is: 19.104803455993533 The number of items in train is: 25 The loss for epoch 9 0.7641921382397413 2 Data saved to: 28_May_202006_27PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_27PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 495.774384
66 2020-04-22 sub_region ... 66 495.760284
67 2020-04-23 sub_region ... 67 495.068512
68 2020-04-24 sub_region ... 68 496.070099
69 2020-04-25 sub_region ... 69 497.589020
70 2020-04-26 sub_region ... 70 495.803619
71 2020-04-27 sub_region ... 71 500.093933
72 2020-04-28 sub_region ... 72 495.899109
73 2020-04-29 sub_region ... 73 495.294128
74 2020-04-30 sub_region ... 74 495.290405
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/media/plotly/test_plot_20_b4b150f6.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 2g463y10
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182744-2g463y10/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 2g463y10 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: jnu41qmm with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 2 use_mask: True wandb: Agent Started Run: jnu41qmm
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/jnu41qmm INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmpudTQxcW1tOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/requirements.txt
The running loss is: 22.70665431022644 The number of items in train is: 25 The loss for epoch 0 0.9082661724090576
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/media/graph/graph_0_summary_7d609e50.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/media
The running loss is: 20.07255493104458 The number of items in train is: 25 The loss for epoch 1 0.8029021972417831
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-history.jsonl
The running loss is: 19.82785291969776 The number of items in train is: 25 The loss for epoch 2 0.7931141167879104 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-history.jsonl
The running loss is: 21.11555242538452 The number of items in train is: 25 The loss for epoch 3 0.8446220970153808
INFO:wandb.wandb_agent:Running runs: ['jnu41qmm'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-history.jsonl
The running loss is: 19.238652385771275 The number of items in train is: 25 The loss for epoch 4 0.769546095430851 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-history.jsonl
The running loss is: 19.169510439038277 The number of items in train is: 25 The loss for epoch 5 0.766780417561531 2 The running loss is: 19.454397417604923 The number of items in train is: 25 The loss for epoch 6 0.778175896704197 3 Stopping model now Data saved to:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-summary.json
28_May_202006_28PM.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-history.jsonl DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_28PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 386.295502
66 2020-04-22 sub_region ... 66 386.368958
67 2020-04-23 sub_region ... 67 386.170776
68 2020-04-24 sub_region ... 68 386.041290
69 2020-04-25 sub_region ... 69 386.396393
70 2020-04-26 sub_region ... 70 386.061401
71 2020-04-27 sub_region ... 71 386.280853
72 2020-04-28 sub_region ... 72 386.269623
73 2020-04-29 sub_region ... 73 386.305389
74 2020-04-30 sub_region ... 74 386.221985
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: jnu41qmm
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/media/plotly/test_plot_14_416120dd.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/media/plotly/test_plot_all_15_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182807-jnu41qmm/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: jnu41qmm INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: le9gifcu with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 2 use_mask: False wandb: Agent Started Run: le9gifcu
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/le9gifcu INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmxlOWdpZmN1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/media/graph/graph_0_summary_953bc13a.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/media
The running loss is: 22.725555673241615 The number of items in train is: 25 The loss for epoch 0 0.9090222269296646
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-history.jsonl
The running loss is: 20.579730972647667 The number of items in train is: 25 The loss for epoch 1 0.8231892389059067
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-history.jsonl
The running loss is: 19.793140269815922 The number of items in train is: 25 The loss for epoch 2 0.7917256107926369 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-history.jsonl
The running loss is: 20.604701928794384 The number of items in train is: 25 The loss for epoch 3 0.8241880771517753
INFO:wandb.wandb_agent:Running runs: ['le9gifcu']
The running loss is: 19.242981515824795 The number of items in train is: 25 The loss for epoch 4 0.7697192606329918 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-history.jsonl
The running loss is: 19.15913762152195 The number of items in train is: 25 The loss for epoch 5 0.766365504860878 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-history.jsonl
The running loss is: 19.610484287142754 The number of items in train is: 25 The loss for epoch 6 0.7844193714857102 3 Stopping model now Data saved to: 28_May_202006_28PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-history.jsonl
Data saved to: 28_May_202006_28PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 408.607574
66 2020-04-22 sub_region ... 66 408.763275
67 2020-04-23 sub_region ... 67 408.592804
68 2020-04-24 sub_region ... 68 408.707031
69 2020-04-25 sub_region ... 69 408.915588
70 2020-04-26 sub_region ... 70 408.787262
71 2020-04-27 sub_region ... 71 409.198151
72 2020-04-28 sub_region ... 72 408.587250
73 2020-04-29 sub_region ... 73 408.494568
74 2020-04-30 sub_region ... 74 408.603363
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: le9gifcu
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/media/plotly/test_plot_14_755375b7.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/media/plotly/test_plot_all_15_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182824-le9gifcu/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: le9gifcu INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: p93c9oel with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 3 use_mask: True wandb: Agent Started Run: p93c9oel
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/p93c9oel INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnA5M2M5b2VsOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/media/graph/graph_0_summary_02eaa5ab.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/media
The running loss is: 19.80333824455738 The number of items in train is: 24 The loss for epoch 0 0.8251390935232242
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-summary.json
The running loss is: 22.39603229612112 The number of items in train is: 24 The loss for epoch 1 0.93316801233838
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-summary.json
The running loss is: 14.272497054189444 The number of items in train is: 24 The loss for epoch 2 0.5946873772578934 1 The running loss is: 14.552330110222101 The number of items in train is: 24 The loss for epoch 3 0.6063470879259208
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-summary.json
The running loss is: 13.988487225025892 The number of items in train is:
INFO:wandb.wandb_agent:Running runs: ['p93c9oel']
24 The loss for epoch 4 0.5828536343760788
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-summary.json
The running loss is: 14.61138915270567 The number of items in train is: 24 The loss for epoch 5 0.6088078813627362 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-summary.json
The running loss is: 13.06570915877819 The number of items in train is: 24 The loss for epoch 6 0.5444045482824246 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-summary.json
The running loss is: 10.021730467677116 The number of items in train is: 24 The loss for epoch 7 0.41757210281987983
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-summary.json
The running loss is: 11.953084118664265 The number of items in train is: 24 The loss for epoch 8 0.49804517161101103 1 The running loss is: 10.495671391487122 The number of items in train is: 24 The loss for epoch 9 0.4373196413119634 2 Data saved to: 28_May_202006_28PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-summary.json
Data saved to: 28_May_202006_28PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 705.102661
66 2020-04-22 sub_region ... 66 705.103027
67 2020-04-23 sub_region ... 67 705.096924
68 2020-04-24 sub_region ... 68 705.107422
69 2020-04-25 sub_region ... 69 705.110229
70 2020-04-26 sub_region ... 70 705.101257
71 2020-04-27 sub_region ... 71 705.131958
72 2020-04-28 sub_region ... 72 705.105957
73 2020-04-29 sub_region ... 73 705.100586
74 2020-04-30 sub_region ... 74 705.097656
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/media/plotly/test_plot_20_2b3d8713.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: p93c9oel
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182844-p93c9oel/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: p93c9oel INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 6g3htrhm with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 3 use_mask: False wandb: Agent Started Run: 6g3htrhm
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/6g3htrhm INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjZnM2h0cmhtOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/media/graph/graph_0_summary_8c905f4c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/media
The running loss is: 19.914355665445328 The number of items in train is: 24 The loss for epoch 0 0.8297648193935553
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-history.jsonl
The running loss is: 22.440757513046265 The number of items in train is: 24 The loss for epoch 1 0.9350315630435944
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-history.jsonl
The running loss is: 14.276884399354458 The number of items in train is: 24 The loss for epoch 2 0.5948701833064357 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-history.jsonl
The running loss is: 14.543040201067924 The number of items in train is: 24 The loss for epoch 3 0.6059600083778302
INFO:wandb.wandb_agent:Running runs: ['6g3htrhm']
The running loss is: 14.132904570549726 The number of items in train is: 24 The loss for epoch 4 0.5888710237729052
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-history.jsonl
The running loss is: 13.135652396827936 The number of items in train is: 24 The loss for epoch 5 0.5473188498678306
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-history.jsonl
The running loss is: 12.404533512890339 The number of items in train is: 24 The loss for epoch 6 0.5168555630370975 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-history.jsonl
The running loss is: 12.101209737360477 The number of items in train is: 24 The loss for epoch 7 0.5042170723900199
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-history.jsonl
The running loss is: 14.456808343529701 The number of items in train is: 24 The loss for epoch 8 0.6023670143137375 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-history.jsonl
The running loss is: 8.79074577242136 The number of items in train is: 24 The loss for epoch 9 0.36628107385089 2 Data saved to: 28_May_202006_29PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_29PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-summary.json
CSV Path below United_States__California__Los_Angeles_County.csv
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-history.jsonl
torch.Size([1, 15, 9]) Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/config.yaml /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 748.224915
66 2020-04-22 sub_region ... 66 748.223572
67 2020-04-23 sub_region ... 67 748.228333
68 2020-04-24 sub_region ... 68 748.199829
69 2020-04-25 sub_region ... 69 748.202820
70 2020-04-26 sub_region ... 70 748.207642
71 2020-04-27 sub_region ... 71 748.156372
72 2020-04-28 sub_region ... 72 748.215271
73 2020-04-29 sub_region ... 73 748.225708
74 2020-04-30 sub_region ... 74 748.218994
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/media/plotly/test_plot_20_62c51e74.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 6g3htrhm
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182906-6g3htrhm/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 6g3htrhm INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: kxdboz50 with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 4 use_mask: True wandb: Agent Started Run: kxdboz50
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/kxdboz50 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmt4ZGJvejUwOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/media/graph/graph_0_summary_507531e2.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/media/graph
The running loss is: 22.587218046188354 The number of items in train is: 24 The loss for epoch 0 0.9411340852578481
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-history.jsonl
The running loss is: 19.35363259911537 The number of items in train is: 24 The loss for epoch 1 0.8064013582964739
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-history.jsonl
The running loss is: 16.189658485352993 The number of items in train is: 24 The loss for epoch 2 0.6745691035563747 The running loss is: 15.560340829193592 The number of items in train is: 24 The loss for epoch 3 0.648347534549733
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-history.jsonl
The running loss is: 15.733900420367718 The number of items in train is: 24 The loss for epoch 4 0.6555791841819882
INFO:wandb.wandb_agent:Running runs: ['kxdboz50']
1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-history.jsonl
The running loss is: 15.708370111882687 The number of items in train is: 24 The loss for epoch 5 0.6545154213284453
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-history.jsonl
The running loss is: 15.470614947378635 The number of items in train is: 24 The loss for epoch 6 0.6446089561407765
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-history.jsonl
The running loss is: 15.394039362668991 The number of items in train is: 24 The loss for epoch 7 0.6414183067778746 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-history.jsonl
The running loss is: 15.352656692266464 The number of items in train is: 24 The loss for epoch 8 0.639694028844436 The running loss is: 15.739529885351658 The number of items in train is: 24 The loss for epoch 9 0.6558137452229857 1 Data saved to:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-summary.json
28_May_202006_29PM.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-history.jsonl DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_29PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 404.308929
66 2020-04-22 sub_region ... 66 405.262390
67 2020-04-23 sub_region ... 67 405.003235
68 2020-04-24 sub_region ... 68 406.038116
69 2020-04-25 sub_region ... 69 403.401398
70 2020-04-26 sub_region ... 70 406.035461
71 2020-04-27 sub_region ... 71 402.491180
72 2020-04-28 sub_region ... 72 404.122070
73 2020-04-29 sub_region ... 73 404.753174
74 2020-04-30 sub_region ... 74 405.485504
[25 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/media/plotly/test_plot_20_dcd65458.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: kxdboz50
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182929-kxdboz50/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: kxdboz50 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: ppzooyba with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 4 use_mask: False wandb: Agent Started Run: ppzooyba
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/ppzooyba INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnBwem9veWJhOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/media/graph/graph_0_summary_a7d93b40.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/media/graph
The running loss is: 22.53992810845375 The number of items in train is: 24 The loss for epoch 0 0.939163671185573
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-history.jsonl
The running loss is: 19.258116833865643 The number of items in train is: 24 The loss for epoch 1 0.8024215347444018
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-history.jsonl
The running loss is: 16.27618619799614 The number of items in train is: 24 The loss for epoch 2 0.6781744249165058 The running loss is: 15.538168050348759 The number of items in train is: 24 The loss for epoch 3 0.6474236687645316
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-history.jsonl INFO:wandb.wandb_agent:Running runs: ['ppzooyba']
The running loss is: 15.75120346993208 The number of items in train is: 24 The loss for epoch 4 0.6563001445805033 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-history.jsonl
The running loss is: 15.720529459416866 The number of items in train is: 24 The loss for epoch 5 0.6550220608090361
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-history.jsonl
The running loss is: 15.435486644506454 The number of items in train is: 24 The loss for epoch 6 0.6431452768544356
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-history.jsonl
The running loss is: 15.386884167790413 The number of items in train is: 24 The loss for epoch 7 0.6411201736579338 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-history.jsonl
The running loss is: 15.28485395014286 The number of items in train is: 24 The loss for epoch 8 0.6368689145892859 The running loss is: 15.743594877421856 The number of items in train is: 24 The loss for epoch 9 0.6559831198925773
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-history.jsonl
1 Data saved to: 28_May_202006_30PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_30PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 400.785828
66 2020-04-22 sub_region ... 66 401.776733
67 2020-04-23 sub_region ... 67 401.619812
68 2020-04-24 sub_region ... 68 403.111847
69 2020-04-25 sub_region ... 69 400.459259
70 2020-04-26 sub_region ... 70 402.953522
71 2020-04-27 sub_region ... 71 400.775208
72 2020-04-28 sub_region ... 72 400.355286
73 2020-04-29 sub_region ... 73 400.579224
74 2020-04-30 sub_region ... 74 401.702911
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-history.jsonl
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/media/plotly/test_plot_20_42d1f4aa.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/media/plotly
wandb: Agent Finished Run: ppzooyba
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_182951-ppzooyba/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: ppzooyba INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: tbmo4dqz with config: batch_size: 2 forecast_history: 15 lr: 0.002 number_encoder_layers: 3 out_seq_length: 5 use_mask: True wandb: Agent Started Run: tbmo4dqz
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/tbmo4dqz INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnRibW80ZHF6OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.002
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 5
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.002
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/media/graph/graph_0_summary_ef7f5c69.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/media/graph
The running loss is: 19.067732363939285 The number of items in train is: 23 The loss for epoch 0 0.8290318419104037
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-history.jsonl
The running loss is: 18.485696509480476 The number of items in train is: 23 The loss for epoch 1 0.8037259351948033 The running loss is: 13.346832662820816 The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-summary.json
23 The loss for epoch 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-history.jsonl
0.5802970722965572 The running loss is: 13.152954459190369 The number of items in train is: 23 The loss for epoch 3 0.5718675851821899 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-history.jsonl
The running loss is: 11.9220210313797 The number of items in train is: 23 The loss for epoch 4 0.5183487404947695
INFO:wandb.wandb_agent:Running runs: ['tbmo4dqz'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-history.jsonl
The running loss is: 8.841944128274918 The number of items in train is: 23 The loss for epoch 5 0.38443235340325727 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-history.jsonl
The running loss is: 11.500840276479721 The number of items in train is: 23 The loss for epoch 6 0.5000365337599879 The running loss is: 7.89447258785367 The number of items in train is: 23 The loss for epoch 7 0.3432379386023335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-history.jsonl
1 The running loss is: 7.601438160985708 The number of items in train is: 23 The loss for epoch 8 0.3304973113472047 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-history.jsonl
The running loss is: 4.348183668218553 The number of items in train is: 23 The loss for epoch 9 0.18905146383558927 Data saved to: 28_May_202006_30PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-history.jsonl
Data saved to: 28_May_202006_30PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 752.961365
66 2020-04-22 sub_region ... 66 752.971558
67 2020-04-23 sub_region ... 67 752.971436
68 2020-04-24 sub_region ... 68 752.982910
69 2020-04-25 sub_region ... 69 753.031128
70 2020-04-26 sub_region ... 70 753.011536
71 2020-04-27 sub_region ... 71 753.077209
72 2020-04-28 sub_region ... 72 752.963745
73 2020-04-29 sub_region ... 73 752.952332
74 2020-04-30 sub_region ... 74 752.973022
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: tbmo4dqz
INFO:wandb.run_manager:shutting down system stats and metadata service
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-events.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/media/plotly/test_plot_all_21_b4a9c644.plotly.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/media/plotly/test_plot_20_b60ea359.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/media/plotly
INFO:wandb.run_manager:stopping streaming files and file change observer
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183009-tbmo4dqz/wandb-metadata.json
INFO:wandb.wandb_agent:Cleaning up finished run: tbmo4dqz
wandb: Network error resolved after 0:00:13.296214, resuming normal operation.
INFO:wandb.wandb_agent:Agent received command: run
INFO:wandb.wandb_agent:Agent starting run with config:
batch_size: 2
forecast_history: 15
lr: 0.0001
number_encoder_layers: 1
out_seq_length: 1
use_mask: True
DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 6mksv9sd with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 1 use_mask: True wandb: Agent Started Run: 6mksv9sd
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/6mksv9sd INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjZta3N2OXNkOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/config.yaml INFO:wandb.wandb_agent:Running runs: ['6mksv9sd'] ERROR:wandb.run_manager:Failed to connect to W&B servers after 10 seconds. Letting user process proceed while attempting to reconnect.
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
13.29942782886792
The number of items in train is:
25
The loss for epoch 0
0.5319771131547167
The running loss is:
15.958505789749324
The number of items in train is:
25
The loss for epoch 1
0.638340231589973
The running loss is:
9.610189565457404
The number of items in train is:
25
The loss for epoch 2
0.38440758261829616
wandb: Network error resolved after 0:00:11.353098, resuming normal operation.
INFO:wandb.run_manager:saving pip packages
INFO:wandb.run_manager:initializing streaming files api
INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-events.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-metadata.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/config.yaml
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/media/graph/graph_0_summary_4695df2b.graph.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/media/graph
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/requirements.txt
The running loss is: 8.331762389920186 The number of items in train is: 25 The loss for epoch 3 0.33327049559680744 1 The running loss is: 6.60916328427993 The number of items in train is: 25 The loss for epoch 4 0.2643665313711972 2 The running loss is: 6.482465127395699 The number of items in train is: 25 The loss for epoch 5 0.25929860509582797
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-summary.json
The running loss is: 7.1515498894732445 The number of items in train is: 25 The loss for epoch 6 0.2860619955789298 The running loss is: 6.881212463777047 The number of items in train is: 25 The loss for epoch 7 0.27524849855108185 The running loss is: 7.631635777652264 The number of items in train is: 25 The loss for epoch 8 0.30526543110609056 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-summary.json
The running loss is: 6.2557811347651295 The number of items in train is: 25 The loss for epoch 9 0.2502312453906052 Data saved to: 28_May_202006_30PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_30PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 801.962341
66 2020-04-22 sub_region ... 66 824.838013
67 2020-04-23 sub_region ... 67 775.098450
68 2020-04-24 sub_region ... 68 820.728394
69 2020-04-25 sub_region ... 69 858.230591
70 2020-04-26 sub_region ... 70 790.655151
71 2020-04-27 sub_region ... 71 957.766296
72 2020-04-28 sub_region ... 72 793.469360
73 2020-04-29 sub_region ... 73 769.994263
74 2020-04-30 sub_region ... 74 771.788574
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-metadata.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/media/plotly/test_plot_20_302bf663.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 6mksv9sd
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183037-6mksv9sd/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 6mksv9sd INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 1na747h3 with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 1 use_mask: False wandb: Agent Started Run: 1na747h3
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/1na747h3 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjFuYTc0N2gzOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
13.510150267975405
The number of items in train is:
25
The loss for epoch 0
0.5404060107190162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/media/graph/graph_0_summary_4bbab05e.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/media
The running loss is: 16.415556019172072 The number of items in train is: 25 The loss for epoch 1 0.6566222407668829 1 The running loss is: 8.791841979138553 The number of items in train is: 25 The loss for epoch 2 0.35167367916554215 The running loss is: 9.764055475010537 The number of items in train is: 25 The loss for epoch 3 0.3905622190004215
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-summary.json
The running loss is: 7.065425713779405 The number of items in train is: 25 The loss for epoch 4 0.2826170285511762 1 The running loss is: 6.248815768616623 The number of items in train is: 25 The loss for epoch 5 0.24995263074466492
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-summary.json
The running loss is: 6.674299073871225 The number of items in train is: 25 The loss for epoch 6 0.266971962954849 1 The running loss is: 7.030096196685918 The number of items in train is: 25 The loss for epoch 7 0.2812038478674367 The running loss is: 8.11010273359716 The number of items in train is: 25 The loss for epoch 8 0.32440410934388636 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-summary.json
The running loss is: 8.296146783512086 The number of items in train is: 25 The loss for epoch 9 0.33184587134048343 Data saved to: 28_May_202006_31PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_31PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv
INFO:wandb.wandb_agent:Running runs: ['1na747h3']
torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 725.814941
66 2020-04-22 sub_region ... 66 739.527832
67 2020-04-23 sub_region ... 67 716.998169
68 2020-04-24 sub_region ... 68 738.462158
69 2020-04-25 sub_region ... 69 756.830688
70 2020-04-26 sub_region ... 70 728.941650
71 2020-04-27 sub_region ... 71 807.169800
72 2020-04-28 sub_region ... 72 725.725586
73 2020-04-29 sub_region ... 73 714.828308
74 2020-04-30 sub_region ... 74 716.417358
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 1na747h3
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/media/plotly/test_plot_20_f6cc8bc8.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183105-1na747h3/wandb-events.jsonl INFO:wandb.wandb_agent:Cleaning up finished run: 1na747h3 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: t27i38bn with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 2 use_mask: True wandb: Agent Started Run: t27i38bn
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/t27i38bn INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnQyN2kzOGJuOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
19.00246888399124
The number of items in train is:
25
The loss for epoch 0
0.7600987553596497
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/media/graph/graph_0_summary_cb13d26f.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/media
The running loss is: 16.036757367663085 The number of items in train is: 25 The loss for epoch 1 0.6414702947065234 1 The running loss is: 10.140504654031247 The number of items in train is: 25 The loss for epoch 2 0.4056201861612499 The running loss is: 11.365801957435906 The number of items in train is: 25 The loss for epoch 3 0.45463207829743624
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-summary.json
1 The running loss is: 15.48135045915842 The number of items in train is: 25 The loss for epoch 4 0.6192540183663369 2 The running loss is: 8.226837230846286 The number of items in train is: 25 The loss for epoch 5 0.32907348923385144
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-history.jsonl
The running loss is: 7.584667568095028 The number of items in train is:
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-summary.json
25 The loss for epoch 6 0.30338670272380114 1 The running loss is: 8.201896103098989 The number of items in train is: 25 The loss for epoch 7 0.3280758441239595 2 The running loss is: 8.102215321734548 The number of items in train is: 25 The loss for epoch 8 0.3240886128693819
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-summary.json
The running loss is: 8.849566355347633 The number of items in train is: 25 The loss for epoch 9 0.35398265421390535 1 Data saved to: 28_May_202006_31PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_31PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['t27i38bn'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 721.418884
66 2020-04-22 sub_region ... 66 729.418823
67 2020-04-23 sub_region ... 67 704.071716
68 2020-04-24 sub_region ... 68 698.017456
69 2020-04-25 sub_region ... 69 721.563843
70 2020-04-26 sub_region ... 70 702.487244
71 2020-04-27 sub_region ... 71 729.263550
72 2020-04-28 sub_region ... 72 726.579834
73 2020-04-29 sub_region ... 73 722.222656
74 2020-04-30 sub_region ... 74 718.942139
[25 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/media/plotly/test_plot_20_359c7a7b.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: t27i38bn
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183122-t27i38bn/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: t27i38bn INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: r7sxv2qw with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 2 use_mask: False wandb: Agent Started Run: r7sxv2qw
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/r7sxv2qw INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnI3c3h2MnF3OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
18.724818095564842
The number of items in train is:
25
The loss for epoch 0
0.7489927238225937
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/media/graph/graph_0_summary_2ebd3a78.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/media
The running loss is: 15.727116253226995 The number of items in train is: 25 The loss for epoch 1 0.6290846501290798 1 The running loss is: 8.56138514901977 The number of items in train is: 25 The loss for epoch 2 0.3424554059607908 The running loss is: 10.345093304291368 The number of items in train is: 25 The loss for epoch 3 0.4138037321716547
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-history.jsonl
1 The running loss is: 10.967490065842867 The number of items in train is: 25 The loss for epoch 4 0.43869960263371466 The running loss is: 7.611721908673644 The number of items in train is: 25 The loss for epoch 5 0.30446887634694575 The running loss is: 7.797869371250272 The number of items in train is: 25 The loss for epoch 6 0.31191477485001085
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-history.jsonl
The running loss is: 7.795289715752006 The number of items in train is: 25 The loss for epoch 7 0.31181158863008024 The running loss is: 6.245977671816945 The number of items in train is: 25 The loss for epoch 8 0.2498391068726778 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-history.jsonl
The running loss is: 6.67815606854856 The number of items in train is: 25 The loss for epoch 9 0.2671262427419424 Data saved to: 28_May_202006_31PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_31PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['r7sxv2qw']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-history.jsonl /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 648.033325
66 2020-04-22 sub_region ... 66 696.399780
67 2020-04-23 sub_region ... 67 633.036682
68 2020-04-24 sub_region ... 68 599.434021
69 2020-04-25 sub_region ... 69 679.843323
70 2020-04-26 sub_region ... 70 630.065979
71 2020-04-27 sub_region ... 71 673.871094
72 2020-04-28 sub_region ... 72 681.052856
73 2020-04-29 sub_region ... 73 670.992432
74 2020-04-30 sub_region ... 74 661.747437
[25 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/media/plotly/test_plot_20_c2722b2a.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: r7sxv2qw
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183135-r7sxv2qw/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: r7sxv2qw INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: fnjnnh45 with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 3 use_mask: True wandb: Agent Started Run: fnjnnh45
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/fnjnnh45 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmZuam5uaDQ1OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
12.417334482073784
The number of items in train is:
24
The loss for epoch 0
0.5173889367530743
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/media/graph/graph_0_summary_eb031a55.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/media
The running loss is: 14.187367584556341 The number of items in train is: 24 The loss for epoch 1 0.5911403160231808 The running loss is: 8.399394869804382 The number of items in train is: 24 The loss for epoch 2 0.34997478624184925 1 The running loss is: 6.990775217302144 The number of items in train is: 24 The loss for epoch 3 0.29128230072092265
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-history.jsonl
The running loss is: 8.946204421576113 The number of items in train is: 24 The loss for epoch 4 0.37275851756567135 1 The running loss is: 15.735674642026424 The number of items in train is: 24 The loss for epoch 5 0.6556531100844344 The running loss is: 7.7431636694818735 The number of items in train is: 24 The loss for epoch 6 0.32263181956174475
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-history.jsonl
1 The running loss is: 4.328124134801328 The number of items in train is: 24 The loss for epoch 7 0.18033850561672202 2 The running loss is: 3.735455541405827 The number of items in train is: 24 The loss for epoch 8 0.15564398089190945 The running loss is: 3.9662495320662856 The number of items in train is: 24 The loss for epoch 9 0.16526039716942856
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-history.jsonl
1 Data saved to: 28_May_202006_31PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_31PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['fnjnnh45'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 755.517212
66 2020-04-22 sub_region ... 66 771.916016
67 2020-04-23 sub_region ... 67 760.687378
68 2020-04-24 sub_region ... 68 735.305786
69 2020-04-25 sub_region ... 69 728.676880
70 2020-04-26 sub_region ... 70 742.574097
71 2020-04-27 sub_region ... 71 690.534424
72 2020-04-28 sub_region ... 72 759.479492
73 2020-04-29 sub_region ... 73 761.484863
74 2020-04-30 sub_region ... 74 765.487183
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/config.yaml
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/media/plotly/test_plot_20_1d2f2943.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: fnjnnh45
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183147-fnjnnh45/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: fnjnnh45 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 31nb2sbt with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 3 use_mask: False wandb: Agent Started Run: 31nb2sbt
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/31nb2sbt INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjMxbmIyc2J0OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
12.737247206270695
The number of items in train is:
24
The loss for epoch 0
0.5307186335946122
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/media/graph/graph_0_summary_a4148c75.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/requirements.txt
The running loss is:
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/media/graph
12.846482746303082
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/media
The number of items in train is: 24 The loss for epoch 1 0.5352701144292951 The running loss is: 8.131057146936655 The number of items in train is: 24 The loss for epoch 2 0.3387940477890273 1 The running loss is: 5.895240487996489 The number of items in train is: 24 The loss for epoch 3 0.24563502033318704 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-history.jsonl
The running loss is: 8.398755688220263 The number of items in train is: 24 The loss for epoch 4 0.34994815367584425 The running loss is: 16.694155018776655 The number of items in train is: 24 The loss for epoch 5 0.6955897924490273 The running loss is: 9.139933694154024 The number of items in train is: 24 The loss for epoch 6 0.380830570589751 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-history.jsonl
The running loss is: 5.126268687658012 The number of items in train is: 24 The loss for epoch 7 0.21359452865241715 2 The running loss is: 4.945361414458603 The number of items in train is: 24 The loss for epoch 8 0.2060567256024418 3 Stopping model now Data saved to: 28_May_202006_32PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-history.jsonl
Data saved to: 28_May_202006_32PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 596.721497
66 2020-04-22 sub_region ... 66 611.140503
67 2020-04-23 sub_region ... 67 602.298889
68 2020-04-24 sub_region ... 68 592.272522
69 2020-04-25 sub_region ... 69 590.910889
70 2020-04-26 sub_region ... 70 592.348145
71 2020-04-27 sub_region ... 71 565.893555
72 2020-04-28 sub_region ... 72 597.307617
73 2020-04-29 sub_region ... 73 598.141846
74 2020-04-30 sub_region ... 74 604.241028
[25 rows x 28 columns]
INFO:wandb.wandb_agent:Running runs: ['31nb2sbt']
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/media/plotly/test_plot_18_b0d51432.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 31nb2sbt
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/media/plotly/test_plot_all_19_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183204-31nb2sbt/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 31nb2sbt INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 7wcnck5x with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 4 use_mask: True wandb: Agent Started Run: 7wcnck5x
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/7wcnck5x INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjd3Y25jazV4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
15.76520562171936
The number of items in train is:
24
The loss for epoch 0
0.65688356757164
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/media/graph/graph_0_summary_d868da83.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/media/graph
The running loss is: 15.477864772081375 The number of items in train is: 24 The loss for epoch 1 0.6449110321700573 1 The running loss is: 12.414291091263294 The number of items in train is: 24 The loss for epoch 2 0.5172621288026372 2 The running loss is: 7.5504014934413135 The number of items in train is: 24 The loss for epoch 3 0.3146000622267214
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-summary.json
The running loss is: 12.823520310223103 The number of items in train is: 24 The loss for epoch 4 0.5343133462592959 1 The running loss is: 6.759511549025774 The number of items in train is: 24 The loss for epoch 5 0.2816463145427406 The running loss is: 6.30719225294888 The number of items in train is: 24 The loss for epoch 6 0.26279967720620334 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-summary.json
The running loss is: 6.85306416079402 The number of items in train is: 24 The loss for epoch 7 0.28554434003308415 2 The running loss is: 6.9310750886797905 The number of items in train is: 24 The loss for epoch 8 0.2887947953616579 The running loss is: 8.455319091677666 The number of items in train is: 24 The loss for epoch 9 0.3523049621532361 1 Data saved to: 28_May_202006_32PM.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-history.jsonl DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-summary.json
Data saved to: 28_May_202006_32PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['7wcnck5x']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 779.131897
66 2020-04-22 sub_region ... 66 792.555298
67 2020-04-23 sub_region ... 67 780.039795
68 2020-04-24 sub_region ... 68 769.559204
69 2020-04-25 sub_region ... 69 768.763855
70 2020-04-26 sub_region ... 70 775.261108
71 2020-04-27 sub_region ... 71 753.575806
72 2020-04-28 sub_region ... 72 784.202087
73 2020-04-29 sub_region ... 73 783.728699
74 2020-04-30 sub_region ... 74 785.233765
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/media/plotly/test_plot_20_f7c7c22f.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/media/plotly
wandb: Agent Finished Run: 7wcnck5x
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183216-7wcnck5x/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 7wcnck5x INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: flfnr7zc with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 4 use_mask: False wandb: Agent Started Run: flfnr7zc
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/flfnr7zc INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmZsZm5yN3pjOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
15.728306412696838
The number of items in train is:
24
The loss for epoch 0
0.655346100529035
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/media/graph/graph_0_summary_0bf0f215.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/media/graph
The running loss is: 13.534287229180336 The number of items in train is: 24 The loss for epoch 1 0.5639286345491806 1 The running loss is: 10.27218734100461 The number of items in train is: 24 The loss for epoch 2 0.4280078058751921 2 The running loss is: 5.5147014474496245 The number of items in train is: 24 The loss for epoch 3 0.22977922697706768
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-history.jsonl
The running loss is: 9.376355066895485 The number of items in train is: 24 The loss for epoch 4 0.3906814611206452 1 The running loss is: 6.485733915120363 The number of items in train is: 24 The loss for epoch 5 0.27023891313001513 2 The running loss is: 7.979923363775015 The number of items in train is: 24 The loss for epoch 6 0.33249680682395893
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-history.jsonl
The running loss is: 8.143040359020233 The number of items in train is: 24 The loss for epoch 7 0.33929334829250973 1 The running loss is: 5.536451553925872 The number of items in train is: 24 The loss for epoch 8 0.230685481413578 2 The running loss is: 4.550636069383472 The number of items in train is: 24 The loss for epoch 9 0.18960983622431135 Data saved to: 28_May_202006_32PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-history.jsonl
Data saved to: 28_May_202006_32PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['flfnr7zc']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 756.924194
66 2020-04-22 sub_region ... 66 788.874878
67 2020-04-23 sub_region ... 67 767.624146
68 2020-04-24 sub_region ... 68 745.193481
69 2020-04-25 sub_region ... 69 751.710815
70 2020-04-26 sub_region ... 70 755.871094
71 2020-04-27 sub_region ... 71 710.098877
72 2020-04-28 sub_region ... 72 772.029785
73 2020-04-29 sub_region ... 73 776.352173
74 2020-04-30 sub_region ... 74 783.972778
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/media/plotly/test_plot_20_75431aa1.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/media/plotly
wandb: Agent Finished Run: flfnr7zc
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183229-flfnr7zc/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: flfnr7zc INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: kiy0qx6s with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 5 use_mask: True wandb: Agent Started Run: kiy0qx6s
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/kiy0qx6s INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmtpeTBxeDZzOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
12.287862300872803
The number of items in train is:
23
The loss for epoch 0
0.5342548826466436
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/media/graph/graph_0_summary_eb18aaf4.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/media
The running loss is: 13.926748000085354 The number of items in train is: 23 The loss for epoch 1 0.6055107826124067 1 The running loss is: 11.354672580957413 The number of items in train is: 23 The loss for epoch 2 0.4936814165633658 2 The running loss is: 9.702557804062963 The number of items in train is: 23 The loss for epoch 3 0.4218503393070853
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-summary.json
The running loss is: 11.627340912818909 The number of items in train is: 23 The loss for epoch 4 0.5055365614269091 1 The running loss is: 6.340536650270224 The number of items in train is: 23 The loss for epoch 5 0.275675506533488 2 The running loss is: 10.078678345307708 The number of items in train is: 23 The loss for epoch 6 0.4382034063177264
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-summary.json
The running loss is: 8.461791813373566 The number of items in train is: 23 The loss for epoch 7 0.3679039918858072 1 The running loss is: 4.9823848977684975 The number of items in train is: 23 The loss for epoch 8 0.21662543033776077 2 The running loss is: 3.317102179862559 The number of items in train is: 23 The loss for epoch 9 0.14422183390706778 Data saved to: 28_May_202006_32PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-summary.json
Data saved to: 28_May_202006_32PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['kiy0qx6s']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 835.124390
66 2020-04-22 sub_region ... 66 840.772278
67 2020-04-23 sub_region ... 67 832.618164
68 2020-04-24 sub_region ... 68 835.653015
69 2020-04-25 sub_region ... 69 821.924072
70 2020-04-26 sub_region ... 70 833.929749
71 2020-04-27 sub_region ... 71 823.226929
72 2020-04-28 sub_region ... 72 831.859680
73 2020-04-29 sub_region ... 73 828.141541
74 2020-04-30 sub_region ... 74 831.716797
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/media/plotly/test_plot_20_7a7a1d5c.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: kiy0qx6s
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183242-kiy0qx6s/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: kiy0qx6s INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: xhheep9s with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 1 out_seq_length: 5 use_mask: False wandb: Agent Started Run: xhheep9s
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/xhheep9s INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnhoaGVlcDlzOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 1
n_time_series: 9
output_seq_length: 5
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 1
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
The running loss is:
12.292471200227737
The number of items in train is:
23
The loss for epoch 0
0.534455269575119
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/media/graph/graph_0_summary_59344234.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/media/graph
The running loss is: 14.315681867301464 The number of items in train is: 23 The loss for epoch 1 0.6224209507522376 1 The running loss is: 10.289977289736271 The number of items in train is: 23 The loss for epoch 2 0.44739031694505527 2 The running loss is: 7.302962388843298 The number of items in train is: 23 The loss for epoch 3 0.3175201038627521
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-summary.json
The running loss is: 10.639127224683762 The number of items in train is: 23 The loss for epoch 4 0.462570748899294 The running loss is: 5.083698073402047 The number of items in train is: 23 The loss for epoch 5 0.2210303510174803 1 The running loss is: 4.266114007681608 The number of items in train is: 23 The loss for epoch 6 0.1854832177252873 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-summary.json
The running loss is: 3.362240646034479 The number of items in train is: 23 The loss for epoch 7 0.14618437591454256 The running loss is: 5.346718622371554 The number of items in train is: 23 The loss for epoch 8 0.2324660270596328 1 The running loss is: 3.976900838315487 The number of items in train is: 23 The loss for epoch 9 0.17290873210067334 Data saved to: 28_May_202006_32PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-summary.json
Data saved to: 28_May_202006_32PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['xhheep9s']
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 773.226318
66 2020-04-22 sub_region ... 66 785.933838
67 2020-04-23 sub_region ... 67 776.867065
68 2020-04-24 sub_region ... 68 779.012268
69 2020-04-25 sub_region ... 69 770.271240
70 2020-04-26 sub_region ... 70 779.572510
71 2020-04-27 sub_region ... 71 765.994873
72 2020-04-28 sub_region ... 72 776.860962
73 2020-04-29 sub_region ... 73 774.669739
74 2020-04-30 sub_region ... 74 777.345093
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/media/plotly/test_plot_20_b982c2b4.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/media
wandb: Agent Finished Run: xhheep9s
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/media/plotly
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183254-xhheep9s/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: xhheep9s INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: eqtdfgtw with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 1 use_mask: True wandb: Agent Started Run: eqtdfgtw
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/eqtdfgtw INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmVxdGRmZ3R3OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/media/graph/graph_0_summary_7174f2ec.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/media
The running loss is: 10.145409000921063 The number of items in train is: 25 The loss for epoch 0 0.4058163600368425 The running loss is: 25.701497849076986 The number of items in train is: 25 The loss for epoch 1 1.0280599139630795
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-history.jsonl
The running loss is: 20.272816001903266 The number of items in train is: 25 The loss for epoch 2 0.8109126400761306
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-history.jsonl
The running loss is: 19.751347083598375 The number of items in train is: 25 The loss for epoch 3 0.790053883343935 The running loss is: 18.793059289455414 The number of items in train is: 25 The loss for epoch 4 0.7517223715782165 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-history.jsonl
The running loss is: 16.612464647740126 The number of items in train is: 25 The loss for epoch 5 0.664498585909605 2
INFO:wandb.wandb_agent:Running runs: ['eqtdfgtw']
The running loss is: 9.692850414896384 The number of items in train is: 25 The loss for epoch 6 0.38771401659585536
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-history.jsonl
The running loss is: 19.73813858255744 The number of items in train is: 25 The loss for epoch 7 0.7895255433022976
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-history.jsonl
The running loss is: 12.20803095260635 The number of items in train is: 25 The loss for epoch 8 0.488321238104254 1 The running loss is: 11.476102620712481 The number of items in train is: 25 The loss for epoch 9 0.4590441048284993 2 Data saved to: 28_May_202006_33PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_33PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 801.515869
66 2020-04-22 sub_region ... 66 801.592773
67 2020-04-23 sub_region ... 67 801.521545
68 2020-04-24 sub_region ... 68 801.597168
69 2020-04-25 sub_region ... 69 801.738037
70 2020-04-26 sub_region ... 70 801.605835
71 2020-04-27 sub_region ... 71 801.837097
72 2020-04-28 sub_region ... 72 801.528931
73 2020-04-29 sub_region ... 73 801.504639
74 2020-04-30 sub_region ... 74 801.555054
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/media/plotly/test_plot_20_fcc89d57.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/media/plotly
wandb: Agent Finished Run: eqtdfgtw
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183306-eqtdfgtw/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: eqtdfgtw INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: xmhc8vgh with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 1 use_mask: False wandb: Agent Started Run: xmhc8vgh
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/xmhc8vgh INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnhtaGM4dmdoOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/media/graph/graph_0_summary_7b56b8de.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/media/graph
The running loss is: 10.096708084311103 The number of items in train is: 25 The loss for epoch 0 0.40386832337244416 The running loss is: 25.173049704171717 The number of items in train is: 25 The loss for epoch 1 1.0069219881668687
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-summary.json
The running loss is: 20.567884587217122 The number of items in train is: 25 The loss for epoch 2 0.8227153834886849
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-summary.json
The running loss is: 16.932578061707318 The number of items in train is: 25 The loss for epoch 3 0.6773031224682927 1 The running loss is: 11.519542964175344 The number of items in train is: 25 The loss for epoch 4 0.4607817185670137 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-summary.json
The running loss is: 8.808077305089682 The number of items in train is: 25 The loss for epoch 5 0.3523230922035873 The running loss is: 11.38830050965771 The number of items in train is: 25 The loss for epoch 6 0.4555320203863084
INFO:wandb.wandb_agent:Running runs: ['xmhc8vgh']
1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-summary.json
The running loss is: 7.139500390738249 The number of items in train is: 25 The loss for epoch 7 0.28558001562952995 2 The running loss is: 9.381748400162905 The number of items in train is: 25 The loss for epoch 8 0.3752699360065162
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-summary.json
The running loss is: 10.521757567301393 The number of items in train is: 25 The loss for epoch 9 0.4208703026920557 Data saved to: 28_May_202006_33PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-summary.json
Data saved to: 28_May_202006_33PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 761.296265
66 2020-04-22 sub_region ... 66 761.453735
67 2020-04-23 sub_region ... 67 761.306885
68 2020-04-24 sub_region ... 68 761.469482
69 2020-04-25 sub_region ... 69 761.735107
70 2020-04-26 sub_region ... 70 761.236694
71 2020-04-27 sub_region ... 71 761.615662
72 2020-04-28 sub_region ... 72 760.781616
73 2020-04-29 sub_region ... 73 760.612366
74 2020-04-30 sub_region ... 74 760.868896
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: xmhc8vgh
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/media/plotly/test_plot_20_b3c2d6a3.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183323-xmhc8vgh/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: xmhc8vgh INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 6gbpik86 with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 2 use_mask: True wandb: Agent Started Run: 6gbpik86
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/6gbpik86 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjZnYnBpazg2OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/media/graph/graph_0_summary_ed06505c.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/media
The running loss is: 20.584014743566513 The number of items in train is: 25 The loss for epoch 0 0.8233605897426606 The running loss is: 17.932957649230957 The number of items in train is: 25 The loss for epoch 1 0.7173183059692383 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-history.jsonl
The running loss is: 15.548911690013483 The number of items in train is: 25 The loss for epoch 2 0.6219564676005394
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-history.jsonl
The running loss is: 14.730191620066762 The number of items in train is: 25 The loss for epoch 3 0.5892076648026705 The running loss is: 15.559959415346384 The number of items in train is: 25 The loss for epoch 4 0.6223983766138553 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-history.jsonl
The running loss is: 11.396017402410507 The number of items in train is: 25 The loss for epoch 5 0.4558406960964203 2 The running loss is: 9.876246387138963 The number of items in train is: 25
INFO:wandb.wandb_agent:Running runs: ['6gbpik86']
The loss for epoch 6 0.3950498554855585 3 Stopping model now Data saved to: 28_May_202006_33PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-history.jsonl
Data saved to: 28_May_202006_33PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 505.433014
66 2020-04-22 sub_region ... 66 504.062500
67 2020-04-23 sub_region ... 67 504.517761
68 2020-04-24 sub_region ... 68 503.535156
69 2020-04-25 sub_region ... 69 506.940369
70 2020-04-26 sub_region ... 70 505.692200
71 2020-04-27 sub_region ... 71 506.551117
72 2020-04-28 sub_region ... 72 507.229309
73 2020-04-29 sub_region ... 73 507.427246
74 2020-04-30 sub_region ... 74 507.579987
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 6gbpik86
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/media/plotly/test_plot_all_15_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/media/plotly/test_plot_14_c7166984.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183341-6gbpik86/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 6gbpik86 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: hb8ltgdq with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 2 use_mask: False wandb: Agent Started Run: hb8ltgdq
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/hb8ltgdq INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmhiOGx0Z2RxOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/media/graph/graph_0_summary_2c5aade9.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/media
The running loss is: 20.51875665783882 The number of items in train is: 25 The loss for epoch 0 0.8207502663135529 The running loss is: 18.745911203324795 The number of items in train is: 25 The loss for epoch 1 0.7498364481329918 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-summary.json
The running loss is: 14.171599298715591 The number of items in train is: 25 The loss for epoch 2 0.5668639719486237 2 The running loss is: 10.58555408520624 The number of items in train is: 25 The loss for epoch 3 0.4234221634082496
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-summary.json
The running loss is: 17.37247646227479 The number of items in train is: 25 The loss for epoch 4 0.6948990584909915
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-summary.json
The running loss is: 10.633864669129252 The number of items in train is: 25 The loss for epoch 5 0.4253545867651701 1 The running loss is: 9.219570185989141 The number of items in train is: 25 The loss for epoch 6 0.36878280743956565
INFO:wandb.wandb_agent:Running runs: ['hb8ltgdq']
2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-summary.json
The running loss is: 8.460247997194529 The number of items in train is: 25 The loss for epoch 7 0.33840991988778113 3 Stopping model now Data saved to: 28_May_202006_33PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_33PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-summary.json
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 684.219727
66 2020-04-22 sub_region ... 66 684.323364
67 2020-04-23 sub_region ... 67 684.080750
68 2020-04-24 sub_region ... 68 683.885254
69 2020-04-25 sub_region ... 69 685.422363
70 2020-04-26 sub_region ... 70 684.204346
71 2020-04-27 sub_region ... 71 685.634094
72 2020-04-28 sub_region ... 72 684.494141
73 2020-04-29 sub_region ... 73 684.385315
74 2020-04-30 sub_region ... 74 684.536133
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/media/plotly/test_plot_16_e1926f30.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: hb8ltgdq
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/media/plotly/test_plot_all_17_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183354-hb8ltgdq/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: hb8ltgdq INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: xnyz7w5q with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 3 use_mask: True wandb: Agent Started Run: xnyz7w5q
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/xnyz7w5q INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnhueXo3dzVxOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/config.yaml
The running loss is: 18.772414907813072 The number of items in train is: 24 The loss for epoch 0 0.7821839544922113
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/media/graph/graph_0_summary_f646221f.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/media
The running loss is: 19.575059436261654 The number of items in train is: 24 The loss for epoch 1 0.8156274765109023
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-summary.json
The running loss is: 15.316046819090843 The number of items in train is: 24 The loss for epoch 2 0.6381686174621185 The running loss is: 11.947806634008884 The number of items in train is: 24 The loss for epoch 3 0.49782527641703683
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-summary.json
The running loss is: 10.35117731243372 The number of items in train is: 24 The loss for epoch 4 0.43129905468473834 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-summary.json
The running loss is: 11.100783228874207 The number of items in train is: 24 The loss for epoch 5 0.4625326345364253 2 The running loss is: 11.392054236494005 The number of items in train is: 24 The loss for epoch 6 0.4746689265205835
INFO:wandb.wandb_agent:Running runs: ['xnyz7w5q'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-summary.json
The running loss is: 5.173449165187776 The number of items in train is: 24 The loss for epoch 7 0.215560381882824 1 The running loss is: 9.032942300196737 The number of items in train is: 24 The loss for epoch 8 0.3763725958415307
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-summary.json
The running loss is: 19.382783502340317 The number of items in train is: 24 The loss for epoch 9 0.8076159792641798 Data saved to: 28_May_202006_34PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_34PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 515.919434
66 2020-04-22 sub_region ... 66 516.543030
67 2020-04-23 sub_region ... 67 515.078186
68 2020-04-24 sub_region ... 68 516.592896
69 2020-04-25 sub_region ... 69 515.011414
70 2020-04-26 sub_region ... 70 514.803894
71 2020-04-27 sub_region ... 71 516.728943
72 2020-04-28 sub_region ... 72 515.620483
73 2020-04-29 sub_region ... 73 515.851013
74 2020-04-30 sub_region ... 74 514.158508
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/media/plotly/test_plot_20_14ad17d9.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: xnyz7w5q
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183412-xnyz7w5q/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: xnyz7w5q INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: p15mr0mp with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 3 use_mask: False wandb: Agent Started Run: p15mr0mp
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/p15mr0mp INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnAxNW1yMG1wOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/media/graph/graph_0_summary_c73f39c4.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/media/graph
The running loss is: 18.797405660152435 The number of items in train is: 24 The loss for epoch 0 0.7832252358396848 The running loss is: 19.402401946485043 The number of items in train is: 24 The loss for epoch 1 0.8084334144368768
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-history.jsonl
The running loss is: 14.805835217237473 The number of items in train is: 24 The loss for epoch 2 0.616909800718228 The running loss is: 9.891540326178074 The number of items in train is: 24 The loss for epoch 3 0.4121475135907531 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-history.jsonl
The running loss is: 5.6078918017446995 The number of items in train is: 24 The loss for epoch 4 0.23366215840602914 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-history.jsonl
The running loss is: 4.041457449435256 The number of items in train is: 24 The loss for epoch 5 0.16839406039313567 3 Stopping model now Data saved to: 28_May_202006_34PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_34PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['p15mr0mp'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 378.001099
66 2020-04-22 sub_region ... 66 379.296204
67 2020-04-23 sub_region ... 67 378.866272
68 2020-04-24 sub_region ... 68 383.898071
69 2020-04-25 sub_region ... 69 369.884155
70 2020-04-26 sub_region ... 70 381.321381
71 2020-04-27 sub_region ... 71 371.926025
72 2020-04-28 sub_region ... 72 378.183258
73 2020-04-29 sub_region ... 73 379.500549
74 2020-04-30 sub_region ... 74 378.237946
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/media/plotly/test_plot_12_9498e4f6.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: p15mr0mp
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/media/plotly/test_plot_all_13_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183429-p15mr0mp/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: p15mr0mp INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 4674polk with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 4 use_mask: True wandb: Agent Started Run: 4674polk
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/4674polk INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjQ2NzRwb2xrOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/media/graph/graph_0_summary_6c7dd0eb.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/media/graph
The running loss is: 21.451373293995857 The number of items in train is: 24 The loss for epoch 0 0.8938072205831608 The running loss is: 18.959119454026222 The number of items in train is: 24 The loss for epoch 1 0.7899633105844259
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-history.jsonl
The running loss is: 16.811535373330116 The number of items in train is: 24 The loss for epoch 2 0.7004806405554215 The running loss is: 15.114799819886684 The number of items in train is: 24 The loss for epoch 3 0.6297833258286119 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-history.jsonl
The running loss is: 13.837378360331059 The number of items in train is: 24 The loss for epoch 4 0.5765574316804608 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-history.jsonl
The running loss is: 10.938715167343616 The number of items in train is: 24 The loss for epoch 5 0.45577979863931734 3 Stopping model now Data saved to: 28_May_202006_34PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_34PM_model.pth
INFO:wandb.wandb_agent:Running runs: ['4674polk'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 379.669922
66 2020-04-22 sub_region ... 66 381.567261
67 2020-04-23 sub_region ... 67 381.375244
68 2020-04-24 sub_region ... 68 386.184875
69 2020-04-25 sub_region ... 69 368.961060
70 2020-04-26 sub_region ... 70 383.574829
71 2020-04-27 sub_region ... 71 370.266907
72 2020-04-28 sub_region ... 72 379.182159
73 2020-04-29 sub_region ... 73 380.702667
74 2020-04-30 sub_region ... 74 380.433716
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/media/plotly/test_plot_12_ab9062b5.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 4674polk
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/media/plotly/test_plot_all_13_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183441-4674polk/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 4674polk INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: m35p2vlx with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 4 use_mask: False wandb: Agent Started Run: m35p2vlx
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/m35p2vlx INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOm0zNXAydmx4OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/media/graph/graph_0_summary_3088cdae.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-events.jsonl
The running loss is: 21.45643301308155 The number of items in train is: 24
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/media/graph
The loss for epoch 0
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/media
0.8940180422117313 The running loss is: 18.878746390342712 The number of items in train is: 24 The loss for epoch 1 0.7866144329309464
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-history.jsonl
The running loss is: 16.909177042543888 The number of items in train is: 24 The loss for epoch 2 0.7045490434393287 The running loss is: 14.992017075419426 The number of items in train is: 24 The loss for epoch 3 0.6246673781424761 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-history.jsonl
The running loss is: 13.057133480906487 The number of items in train is: 24 The loss for epoch 4 0.5440472283711036 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-history.jsonl
The running loss is: 7.168162252753973 The number of items in train is: 24 The loss for epoch 5 0.2986734271980822 3 Stopping model now Data saved to: 28_May_202006_35PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_35PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
INFO:wandb.wandb_agent:Running runs: ['m35p2vlx']
CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 381.720642
66 2020-04-22 sub_region ... 66 383.487915
67 2020-04-23 sub_region ... 67 382.618774
68 2020-04-24 sub_region ... 68 386.354279
69 2020-04-25 sub_region ... 69 371.824524
70 2020-04-26 sub_region ... 70 384.303436
71 2020-04-27 sub_region ... 71 373.255554
72 2020-04-28 sub_region ... 72 381.710022
73 2020-04-29 sub_region ... 73 383.322937
74 2020-04-30 sub_region ... 74 382.421906
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-history.jsonl
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-summary.json
wandb: Agent Finished Run: m35p2vlx
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/media/plotly/test_plot_all_13_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/media/plotly/test_plot_12_d4e101e3.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/media/plotly INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183459-m35p2vlx/wandb-events.jsonl INFO:wandb.wandb_agent:Cleaning up finished run: m35p2vlx INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 5 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 0czseyt9 with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 5 use_mask: True wandb: Agent Started Run: 0czseyt9
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/0czseyt9 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjBjenNleXQ5OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/media/graph/graph_0_summary_8186c48a.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-events.jsonl
The running loss is:
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/media/graph
17.729351297020912 The number of items in train is: 23 The loss for epoch 0 0.7708413607400396 The running loss is: 17.55451713502407 The number of items in train is: 23 The loss for epoch 1 0.7632398754358292
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-summary.json
The running loss is: 13.769226357340813 The number of items in train is: 23 The loss for epoch 2 0.5986620155365571 1 The running loss is: 10.689910009503365 The number of items in train is: 23 The loss for epoch 3 0.46477869606536365 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-summary.json
The running loss is: 10.719441205263138 The number of items in train is: 23 The loss for epoch 4 0.4660626610983973 The running loss is: 11.102143481373787 The number of items in train is: 23 The loss for epoch 5 0.48270189049451245 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-summary.json
The running loss is: 8.552160829305649 The number of items in train is: 23 The loss for epoch 6 0.3718330795350282
INFO:wandb.wandb_agent:Running runs: ['0czseyt9'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-summary.json
The running loss is: 8.911982571706176 The number of items in train is: 23 The loss for epoch 7 0.3874775031176598 1 The running loss is: 10.839222326874733 The number of items in train is: 23 The loss for epoch 8 0.4712705359510753 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-summary.json
The running loss is: 6.471042025834322 The number of items in train is: 23 The loss for epoch 9 0.28134965329714445 Data saved to: 28_May_202006_35PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_35PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 682.042175
66 2020-04-22 sub_region ... 66 684.554626
67 2020-04-23 sub_region ... 67 676.001160
68 2020-04-24 sub_region ... 68 688.336792
69 2020-04-25 sub_region ... 69 666.439087
70 2020-04-26 sub_region ... 70 671.417664
71 2020-04-27 sub_region ... 71 689.934692
72 2020-04-28 sub_region ... 72 674.941956
73 2020-04-29 sub_region ... 73 665.878967
74 2020-04-30 sub_region ... 74 658.025818
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/media/plotly/test_plot_20_790f76c3.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 0czseyt9
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183511-0czseyt9/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 0czseyt9 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 5 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 4usmwcfe with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 2 out_seq_length: 5 use_mask: False wandb: Agent Started Run: 4usmwcfe
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/4usmwcfe INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjR1c213Y2ZlOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 5
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 5
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 2
n_time_series: 9
output_seq_length: 5
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 2
out_seq_length:
desc: null
value: 5
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/config.yaml
The running loss is: 17.749317228794098 The number of items in train is: 23 The loss for epoch 0
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/requirements.txt
0.7717094447301782
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/media/graph/graph_0_summary_df2ed202.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/media/graph
The running loss is: 17.551174700260162 The number of items in train is: 23 The loss for epoch 1 0.7630945521852245
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-history.jsonl
The running loss is: 13.54475101083517 The number of items in train is: 23 The loss for epoch 2 0.5889022178623987 1 The running loss is: 10.344177260994911 The number of items in train is: 23 The loss for epoch 3 0.44974683743456134 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-history.jsonl
The running loss is: 9.028607338666916 The number of items in train is: 23 The loss for epoch 4 0.39254814515943115 3 Stopping model now Data saved to: 28_May_202006_35PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_35PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 391.728210
66 2020-04-22 sub_region ... 66 392.369507
67 2020-04-23 sub_region ... 67 392.830872
68 2020-04-24 sub_region ... 68 397.064850
69 2020-04-25 sub_region ... 69 385.320221
70 2020-04-26 sub_region ... 70 393.841858
71 2020-04-27 sub_region ... 71 388.418610
72 2020-04-28 sub_region ... 72 390.187714
73 2020-04-29 sub_region ... 73 390.757416
74 2020-04-30 sub_region ... 74 391.191467
[25 rows x 28 columns]
INFO:wandb.wandb_agent:Running runs: ['4usmwcfe']
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/media/plotly/test_plot_10_ca77e0e0.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 4usmwcfe
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/media/plotly/test_plot_all_11_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183528-4usmwcfe/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: 4usmwcfe INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 1 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: s7oyw8gr with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 1 use_mask: True wandb: Agent Started Run: s7oyw8gr
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/s7oyw8gr INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnM3b3l3OGdyOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/media/graph/graph_0_summary_9203cfa0.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/media
The running loss is: 10.765136603731662 The number of items in train is: 25 The loss for epoch 0 0.4306054641492665
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-history.jsonl
The running loss is: 26.137935783714056 The number of items in train is: 25 The loss for epoch 1 1.0455174313485622
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-history.jsonl
The running loss is: 22.011536345118657 The number of items in train is: 25 The loss for epoch 2 0.8804614538047463
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-history.jsonl
The running loss is: 18.903898189775646 The number of items in train is: 25 The loss for epoch 3 0.7561559275910258
INFO:wandb.wandb_agent:Running runs: ['s7oyw8gr'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-history.jsonl
The running loss is: 18.865029871463776 The number of items in train is: 25 The loss for epoch 4 0.754601194858551
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-history.jsonl
The running loss is: 19.70023591836798 The number of items in train is: 25 The loss for epoch 5 0.7880094367347192 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-history.jsonl
The running loss is: 18.773742999881506 The number of items in train is: 25 The loss for epoch 6 0.7509497199952603 The running loss is: 19.47792062163353 The number of items in train is: 25
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-summary.json
The loss for epoch 7
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-history.jsonl
0.7791168248653412 The running loss is: 19.640535548329353 The number of items in train is: 25 The loss for epoch 8 0.7856214219331741 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-history.jsonl
The running loss is: 19.046687852591276 The number of items in train is: 25 The loss for epoch 9 0.7618675141036511 Data saved to: 28_May_202006_35PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-history.jsonl
Data saved to: 28_May_202006_35PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 496.016388
66 2020-04-22 sub_region ... 66 496.046783
67 2020-04-23 sub_region ... 67 495.631775
68 2020-04-24 sub_region ... 68 495.853607
69 2020-04-25 sub_region ... 69 496.806030
70 2020-04-26 sub_region ... 70 495.802399
71 2020-04-27 sub_region ... 71 497.801544
72 2020-04-28 sub_region ... 72 496.134644
73 2020-04-29 sub_region ... 73 495.858582
74 2020-04-30 sub_region ... 74 495.841339
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/config.yaml
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/media/plotly/test_plot_20_59b2352c.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/media
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/media/plotly
wandb: Agent Finished Run: s7oyw8gr
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183541-s7oyw8gr/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: s7oyw8gr INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 1 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: e51i13de with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 1 use_mask: False wandb: Agent Started Run: e51i13de
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/e51i13de INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmU1MWkxM2RlOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 1
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 1
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 1
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 1
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/media/graph/graph_0_summary_567e9911.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/media/graph INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/media
The running loss is: 11.082796600530855 The number of items in train is: 25 The loss for epoch 0 0.44331186402123424
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-history.jsonl
The running loss is: 26.767839308828115 The number of items in train is: 25 The loss for epoch 1 1.0707135723531247
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-history.jsonl
The running loss is: 19.91911711357534 The number of items in train is: 25 The loss for epoch 2 0.7967646845430135
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-history.jsonl
The running loss is: 18.481270607560873 The number of items in train is: 25 The loss for epoch 3 0.7392508243024349 1
INFO:wandb.wandb_agent:Running runs: ['e51i13de'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-history.jsonl
The running loss is: 18.93701122701168 The number of items in train is: 25 The loss for epoch 4 0.7574804490804672 The running loss is: 19.590259796008468 The number of items in train is: 25 The loss for epoch 5 0.7836103918403388
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-history.jsonl
The running loss is: 18.897615873254836 The number of items in train is: 25 The loss for epoch 6 0.7559046349301934 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-history.jsonl
The running loss is: 19.6400615721941 The number of items in train is: 25 The loss for epoch 7 0.785602462887764
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-history.jsonl
The running loss is: 19.225644499063492 The number of items in train is: 25 The loss for epoch 8 0.7690257799625396 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-history.jsonl
The running loss is: 19.104803455993533 The number of items in train is: 25 The loss for epoch 9 0.7641921382397413 2 Data saved to: 28_May_202006_36PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_36PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-history.jsonl
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 495.774384
66 2020-04-22 sub_region ... 66 495.760284
67 2020-04-23 sub_region ... 67 495.068512
68 2020-04-24 sub_region ... 68 496.070099
69 2020-04-25 sub_region ... 69 497.589020
70 2020-04-26 sub_region ... 70 495.803619
71 2020-04-27 sub_region ... 71 500.093933
72 2020-04-28 sub_region ... 72 495.899109
73 2020-04-29 sub_region ... 73 495.294128
74 2020-04-30 sub_region ... 74 495.290405
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-summary.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/media/plotly/test_plot_20_b4b150f6.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: e51i13de
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183603-e51i13de/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: e51i13de INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 2 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: b702zpgo with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 2 use_mask: True wandb: Agent Started Run: b702zpgo
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/b702zpgo INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmI3MDJ6cGdvOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/media/graph/graph_0_summary_0b9fdb05.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/media/graph
The running loss is: 22.70665431022644 The number of items in train is: 25 The loss for epoch 0 0.9082661724090576
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-history.jsonl
The running loss is: 20.07255493104458 The number of items in train is: 25 The loss for epoch 1 0.8029021972417831
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-history.jsonl
The running loss is: 19.82785291969776 The number of items in train is: 25 The loss for epoch 2 0.7931141167879104 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-history.jsonl
The running loss is: 21.11555242538452 The number of items in train is: 25 The loss for epoch 3 0.8446220970153808
INFO:wandb.wandb_agent:Running runs: ['b702zpgo']
The running loss is: 19.238652385771275 The number of items in train is: 25 The loss for epoch 4 0.769546095430851
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-history.jsonl
1 The running loss is: 19.169510439038277 The number of items in train is: 25 The loss for epoch 5 0.766780417561531 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-history.jsonl
The running loss is: 19.454397417604923 The number of items in train is: 25 The loss for epoch 6 0.778175896704197 3 Stopping model now Data saved to: 28_May_202006_36PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-history.jsonl
Data saved to: 28_May_202006_36PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 386.295502
66 2020-04-22 sub_region ... 66 386.368958
67 2020-04-23 sub_region ... 67 386.170776
68 2020-04-24 sub_region ... 68 386.041290
69 2020-04-25 sub_region ... 69 386.396393
70 2020-04-26 sub_region ... 70 386.061401
71 2020-04-27 sub_region ... 71 386.280853
72 2020-04-28 sub_region ... 72 386.269623
73 2020-04-29 sub_region ... 73 386.305389
74 2020-04-30 sub_region ... 74 386.221985
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: b702zpgo
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/media/plotly/test_plot_all_15_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/media/plotly/test_plot_14_416120dd.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183626-b702zpgo/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: b702zpgo INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 2 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: vid0ihm1 with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 2 use_mask: False wandb: Agent Started Run: vid0ihm1
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/vid0ihm1 INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOnZpZDBpaG0xOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 2
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 2
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 2
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 2
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/media/graph/graph_0_summary_cd981548.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/media/graph
The running loss is: 22.725555673241615 The number of items in train is: 25 The loss for epoch 0 0.9090222269296646
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-history.jsonl
The running loss is: 20.579730972647667 The number of items in train is: 25 The loss for epoch 1 0.8231892389059067
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-history.jsonl
The running loss is: 19.793140269815922 The number of items in train is: 25 The loss for epoch 2 0.7917256107926369 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-history.jsonl
The running loss is: 20.604701928794384 The number of items in train is: 25 The loss for epoch 3 0.8241880771517753
INFO:wandb.wandb_agent:Running runs: ['vid0ihm1']
The running loss is: 19.242981515824795 The number of items in train is: 25 The loss for epoch 4 0.7697192606329918
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-history.jsonl
1 The running loss is: 19.15913762152195 The number of items in train is: 25 The loss for epoch 5 0.766365504860878 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-history.jsonl
The running loss is: 19.610484287142754 The number of items in train is: 25 The loss for epoch 6 0.7844193714857102 3 Stopping model now Data saved to: 28_May_202006_36PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-history.jsonl
Data saved to: 28_May_202006_36PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 408.607574
66 2020-04-22 sub_region ... 66 408.763275
67 2020-04-23 sub_region ... 67 408.592804
68 2020-04-24 sub_region ... 68 408.707031
69 2020-04-25 sub_region ... 69 408.915588
70 2020-04-26 sub_region ... 70 408.787262
71 2020-04-27 sub_region ... 71 409.198151
72 2020-04-28 sub_region ... 72 408.587250
73 2020-04-29 sub_region ... 73 408.494568
74 2020-04-30 sub_region ... 74 408.603363
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: vid0ihm1
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/media/plotly/test_plot_all_15_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/media/plotly/test_plot_14_755375b7.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183644-vid0ihm1/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: vid0ihm1 INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 3 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: hes2lsqt with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 3 use_mask: True wandb: Agent Started Run: hes2lsqt
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/hes2lsqt INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmhlczJsc3F0OmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/media/graph/graph_0_summary_76c987d5.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/media/graph
The running loss is: 19.80333824455738 The number of items in train is: 24 The loss for epoch 0 0.8251390935232242
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-history.jsonl
The running loss is: 22.39603229612112 The number of items in train is: 24 The loss for epoch 1 0.93316801233838
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-history.jsonl
The running loss is: 14.272497054189444 The number of items in train is: 24 The loss for epoch 2 0.5946873772578934 1 The running loss is: 14.552330110222101 The number of items in train is: 24 The loss for epoch 3 0.6063470879259208
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-history.jsonl
The running loss is: 13.988487225025892 The number of items in train is: 24 The loss for epoch 4 0.5828536343760788
INFO:wandb.wandb_agent:Running runs: ['hes2lsqt'] INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-history.jsonl
The running loss is: 14.61138915270567 The number of items in train is: 24 The loss for epoch 5 0.6088078813627362 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-history.jsonl
The running loss is: 13.06570915877819 The number of items in train is: 24 The loss for epoch 6 0.5444045482824246 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-history.jsonl
The running loss is: 10.021730467677116 The number of items in train is: 24 The loss for epoch 7 0.41757210281987983
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-history.jsonl
The running loss is: 11.953084118664265 The number of items in train is: 24 The loss for epoch 8 0.49804517161101103 1 The running loss is: 10.495671391487122 The number of items in train is: 24 The loss for epoch 9 0.4373196413119634 2 Data saved to: 28_May_202006_37PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-history.jsonl
Data saved to: 28_May_202006_37PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 705.102661
66 2020-04-22 sub_region ... 66 705.103027
67 2020-04-23 sub_region ... 67 705.096924
68 2020-04-24 sub_region ... 68 705.107422
69 2020-04-25 sub_region ... 69 705.110229
70 2020-04-26 sub_region ... 70 705.101257
71 2020-04-27 sub_region ... 71 705.131958
72 2020-04-28 sub_region ... 72 705.105957
73 2020-04-29 sub_region ... 73 705.100586
74 2020-04-30 sub_region ... 74 705.097656
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: hes2lsqt
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/media/plotly/test_plot_20_2b3d8713.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183701-hes2lsqt/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: hes2lsqt INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 3 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: e6va3wtf with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 3 use_mask: False wandb: Agent Started Run: e6va3wtf
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/e6va3wtf INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmU2dmEzd3RmOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 3
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 3
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 3
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 3
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/media/graph/graph_0_summary_25ec0b50.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/media/graph
The running loss is: 19.914355665445328 The number of items in train is: 24 The loss for epoch 0 0.8297648193935553
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-summary.json
The running loss is: 22.440757513046265 The number of items in train is: 24 The loss for epoch 1 0.9350315630435944
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-summary.json
The running loss is: 14.276884399354458 The number of items in train is: 24 The loss for epoch 2 0.5948701833064357 1 The running loss is: 14.543040201067924 The number of items in train is: 24 The loss for epoch 3 0.6059600083778302
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-summary.json INFO:wandb.wandb_agent:Running runs: ['e6va3wtf']
The running loss is: 14.132904570549726 The number of items in train is: 24 The loss for epoch 4 0.5888710237729052
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-summary.json
The running loss is: 13.135652396827936 The number of items in train is: 24 The loss for epoch 5 0.5473188498678306
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-summary.json
The running loss is: 12.404533512890339 The number of items in train is: 24 The loss for epoch 6 0.5168555630370975 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-summary.json
The running loss is: 12.101209737360477 The number of items in train is: 24 The loss for epoch 7 0.5042170723900199
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-summary.json
The running loss is: 14.456808343529701 The number of items in train is: 24 The loss for epoch 8 0.6023670143137375 1 The running loss is: 8.79074577242136 The number of items in train is: 24 The loss for epoch 9 0.36628107385089 2
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-summary.json
Data saved to: 28_May_202006_37PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_37PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/config.yaml
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 748.224915
66 2020-04-22 sub_region ... 66 748.223572
67 2020-04-23 sub_region ... 67 748.228333
68 2020-04-24 sub_region ... 68 748.199829
69 2020-04-25 sub_region ... 69 748.202820
70 2020-04-26 sub_region ... 70 748.207642
71 2020-04-27 sub_region ... 71 748.156372
72 2020-04-28 sub_region ... 72 748.215271
73 2020-04-29 sub_region ... 73 748.225708
74 2020-04-30 sub_region ... 74 748.218994
[25 rows x 28 columns]
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-summary.json
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-history.jsonl
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-summary.json
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/media/plotly/test_plot_20_62c51e74.plotly.json
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: e6va3wtf
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/media/plotly INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183719-e6va3wtf/wandb-metadata.json INFO:wandb.wandb_agent:Cleaning up finished run: e6va3wtf INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 4 use_mask: True DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: 31vcsdvd with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 4 use_mask: True wandb: Agent Started Run: 31vcsdvd
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/31vcsdvd INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOjMxdmNzZHZkOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: true
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: true
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/media/graph/graph_0_summary_1f56b7c7.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/media/graph
The running loss is: 22.587218046188354 The number of items in train is: 24 The loss for epoch 0 0.9411340852578481
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-history.jsonl
The running loss is: 19.35363259911537 The number of items in train is: 24 The loss for epoch 1 0.8064013582964739
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-history.jsonl
The running loss is: 16.189658485352993 The number of items in train is: 24 The loss for epoch 2 0.6745691035563747 The running loss is: 15.560340829193592 The number of items in train is: 24
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-summary.json
The loss for epoch 3
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-history.jsonl
0.648347534549733
INFO:wandb.wandb_agent:Running runs: ['31vcsdvd']
The running loss is: 15.733900420367718 The number of items in train is: 24 The loss for epoch 4 0.6555791841819882 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-history.jsonl
The running loss is: 15.708370111882687 The number of items in train is: 24 The loss for epoch 5 0.6545154213284453
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-history.jsonl
The running loss is: 15.470614947378635 The number of items in train is: 24 The loss for epoch 6 0.6446089561407765
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-history.jsonl
The running loss is: 15.394039362668991 The number of items in train is: 24 The loss for epoch 7 0.6414183067778746 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-history.jsonl
The running loss is: 15.352656692266464 The number of items in train is: 24 The loss for epoch 8 0.639694028844436
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-history.jsonl
The running loss is: 15.739529885351658 The number of items in train is: 24 The loss for epoch 9 0.6558137452229857 1 Data saved to: 28_May_202006_37PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token
Data saved to: 28_May_202006_37PM_model.pth
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/config.yaml
interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 404.308929
66 2020-04-22 sub_region ... 66 405.262390
67 2020-04-23 sub_region ... 67 405.003235
68 2020-04-24 sub_region ... 68 406.038116
69 2020-04-25 sub_region ... 69 403.401398
70 2020-04-26 sub_region ... 70 406.035461
71 2020-04-27 sub_region ... 71 402.491180
72 2020-04-28 sub_region ... 72 404.122070
73 2020-04-29 sub_region ... 73 404.753174
74 2020-04-30 sub_region ... 74 405.485504
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/config.yaml
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0.
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Oblique.ttf) oblique normal 400 normal>) = 1.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralItalic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmss10' (cmss10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmr10' (cmr10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniIta.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUni.ttf) normal normal regular normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymBol.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmb10' (cmb10.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-BoldOblique.ttf) oblique normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Bold.ttf) normal normal 700 condensed>) = 10.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Italic.ttf) italic normal 400 condensed>) = 11.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans' (LiberationSans-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-Regular.ttf) normal normal 400 condensed>) = 10.25
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Bold.ttf) normal normal 700 normal>) = 10.335
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Sans Narrow' (LiberationSansNarrow-BoldItalic.ttf) italic normal 700 condensed>) = 11.535
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Humor Sans' (Humor-Sans.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Italic.ttf) italic normal 400 normal>) = 11.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Serif' (LiberationSerif-Regular.ttf) normal normal 400 normal>) = 10.05
DEBUG:matplotlib.font_manager:findfont: score(<Font 'Liberation Mono' (LiberationMono-BoldItalic.ttf) italic normal 700 normal>) = 11.335
DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0 to DejaVu Sans ('/usr/local/lib/python3.6/dist-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf') with score of 0.050000.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mplexporter/exporter.py:84: UserWarning:
Blended transforms not yet supported. Zoom behavior may not work as expected.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/mpltools.py:368: MatplotlibDeprecationWarning:
The is_frame_like function was deprecated in Matplotlib 3.1 and will be removed in 3.3.
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-summary.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-history.jsonl
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/media/plotly/test_plot_20_dcd65458.plotly.json
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/media
INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/media/plotly
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:410: UserWarning:
Bummer! Plotly can currently only draw Line2D objects from matplotlib that are in 'data' coordinates!
/usr/local/lib/python3.6/dist-packages/plotly/matplotlylib/renderer.py:512: UserWarning:
I found a path object that I don't think is part of a bar chart. Ignoring.
wandb: Agent Finished Run: 31vcsdvd
INFO:wandb.run_manager:shutting down system stats and metadata service INFO:wandb.run_manager:stopping streaming files and file change observer INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-metadata.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-events.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-summary.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/media/plotly/test_plot_all_21_b4a9c644.plotly.json INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183742-31vcsdvd/media/plotly INFO:wandb.wandb_agent:Cleaning up finished run: 31vcsdvd INFO:wandb.wandb_agent:Agent received command: run INFO:wandb.wandb_agent:Agent starting run with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 4 use_mask: False DEBUG:wandb.wandb_config:wandb dir not provided, skipping defaults
wandb: Agent Starting Run: kvnf68wm with config: batch_size: 2 forecast_history: 15 lr: 0.0001 number_encoder_layers: 3 out_seq_length: 4 use_mask: False wandb: Agent Started Run: kvnf68wm
DEBUG:wandb.wandb_config:no defaults not found in config-defaults.yaml
DEBUG:git.cmd:Popen(['git', 'cat-file', '--batch-check'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=<valid stream>) DEBUG:wandb.meta:probe for git information DEBUG:git.cmd:Popen(['git', 'rev-parse', '--show-toplevel'], cwd=/content/github_aistream-peelout_flow-forecast, universal_newlines=False, shell=None, istream=None) DEBUG:wandb.run_manager:Initialized sync for covid_forecast/kvnf68wm INFO:wandb.run_manager:system metrics and metadata threads started INFO:wandb.run_manager:checking resume status, waiting at most 10 seconds INFO:wandb.run_manager:resuming run from id: UnVuOnYxOmt2bmY2OHdtOmNvdmlkX2ZvcmVjYXN0OmNvdmlk INFO:wandb.run_manager:upserting run before process can begin, waiting at most 10 seconds INFO:wandb.run_manager:saving pip packages INFO:wandb.run_manager:initializing streaming files api INFO:wandb.run_manager:unblocking file change observer, beginning sync with W&B servers
sucessfully deleted layers Weights sucessfully loaded interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv
WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable WARNING:google.auth._default:No project ID could be determined. Consider running `gcloud config set project` or setting the GOOGLE_CLOUD_PROJECT environment variable
Using Wandb config:
wandb_version: 1
GCS:
desc: null
value: true
_wandb:
desc: null
value:
cli_version: 0.8.36
framework: torch
is_jupyter_run: true
is_kaggle_kernel: false
python_version: 3.6.9
batch_size:
desc: null
value: 2
dataset_params:
desc: null
value:
batch_size: 2
class: default
forecast_history: 15
forecast_length: 4
interpolate: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaler: StandardScaler
target_col:
- new_cases
test_path: United_States__California__Los_Angeles_County.csv
train_end: 67
training_path: United_States__California__Los_Angeles_County.csv
valid_end: 86
valid_start: 68
validation_path: United_States__California__Los_Angeles_County.csv
early_stopping:
desc: null
value:
patience: 3
forecast_history:
desc: null
value: 15
forward_params:
desc: null
value: {}
inference_params:
desc: null
value:
dataset_params:
file_path: United_States__California__Los_Angeles_County.csv
forecast_history: 15
forecast_length: 4
interpolate_param: false
relevant_cols:
- new_cases
- month
- weekday
- mobility_retail_recreation
- mobility_grocery_pharmacy
- mobility_parks
- mobility_transit_stations
- mobility_workplaces
- mobility_residential
scaling: StandardScaler
target_col:
- new_cases
datetime_start: '2020-04-21'
decoder_params:
decoder_function: simple_decode
unsqueeze_dim: 1
hours_to_forecast: 10
test_csv_path: United_States__California__Los_Angeles_County.csv
lr:
desc: null
value: 0.0001
metrics:
desc: null
value:
- MSE
model_name:
desc: null
value: CustomTransformerDecoder
model_params:
desc: null
value:
n_layers_encoder: 3
n_time_series: 9
output_seq_length: 4
seq_length: 15
use_mask: false
model_type:
desc: null
value: PyTorch
number_encoder_layers:
desc: null
value: 3
out_seq_length:
desc: null
value: 4
sweep:
desc: null
value: true
training_params:
desc: null
value:
batch_size: 2
criterion: MSE
epochs: 10
lr: 0.0001
optim_params: {}
optimizer: Adam
use_mask:
desc: null
value: false
wandb:
desc: null
value: false
weight_path:
desc: null
value: 25_May_202010_29PM_model.pth
weight_path_add:
desc: null
value:
excluded_layers:
- out_length_lay.weight
- out_length_lay.bias
- dense_shape.weight
- dense_shape.bias
Torch is using cpu
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/config.yaml INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-metadata.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-history.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/media/graph/graph_0_summary_0966bf99.graph.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-events.jsonl INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-summary.json INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/requirements.txt INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/media INFO:wandb.run_manager:file/dir created: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/media/graph
The running loss is: 22.53992810845375 The number of items in train is: 24 The loss for epoch 0 0.939163671185573
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-summary.json
The running loss is: 19.258116833865643 The number of items in train is: 24 The loss for epoch 1 0.8024215347444018
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-summary.json
The running loss is: 16.27618619799614 The number of items in train is: 24 The loss for epoch 2 0.6781744249165058 The running loss is: 15.538168050348759 The number of items in train is: 24 The loss for epoch 3 0.6474236687645316
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-summary.json
The running loss is: 15.75120346993208 The number of items in train is: 24 The loss for epoch 4 0.6563001445805033
INFO:wandb.wandb_agent:Running runs: ['kvnf68wm']
1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-summary.json
The running loss is: 15.720529459416866 The number of items in train is: 24 The loss for epoch 5 0.6550220608090361
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-summary.json
The running loss is: 15.435486644506454 The number of items in train is: 24 The loss for epoch 6 0.6431452768544356
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-summary.json
The running loss is: 15.386884167790413 The number of items in train is: 24 The loss for epoch 7 0.6411201736579338 1
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-summary.json
The running loss is: 15.28485395014286 The number of items in train is: 24 The loss for epoch 8 0.6368689145892859 The running loss is: 15.743594877421856 The number of items in train is: 24 The loss for epoch 9 0.6559831198925773 1 Data saved to: 28_May_202006_38PM.json
DEBUG:google.auth.transport.requests:Making request: POST https://oauth2.googleapis.com/token INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-history.jsonl INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-summary.json
Data saved to: 28_May_202006_38PM_model.pth interpolate should be below Now loading and scaling United_States__California__Los_Angeles_County.csv CSV Path below United_States__California__Los_Angeles_County.csv torch.Size([1, 15, 9]) Add debugging crap below
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:96: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'] = 0 /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:97: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df['preds'][history_length:] = end_tensor.numpy().tolist() /usr/local/lib/python3.6/dist-packages/pandas/core/series.py:1042: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self._set_with(key, value) /content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:59: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy df, end_tensor, forecast_history, junk, test_data, prediction_samples = infer_on_torch_model(model, **inference_params)
torch.Size([10]) test_data scale Un-transforming data
/content/github_aistream-peelout_flow-forecast/flood_forecast/evaluator.py:67: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
df['preds'][history_length:] = end_tensor_list
/content/github_aistream-peelout_flow-forecast/flood_forecast/trainer.py:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
test_acc = evaluate_model(trained_model, model_type, params["dataset_params"]["target_col"], params["metrics"], params["inference_params"], {})
Current historical dataframe
date level ... original_index preds
50 2020-04-06 sub_region ... 50 0.000000
51 2020-04-07 sub_region ... 51 0.000000
52 2020-04-08 sub_region ... 52 0.000000
53 2020-04-09 sub_region ... 53 0.000000
54 2020-04-10 sub_region ... 54 0.000000
55 2020-04-11 sub_region ... 55 0.000000
56 2020-04-12 sub_region ... 56 0.000000
57 2020-04-13 sub_region ... 57 0.000000
58 2020-04-14 sub_region ... 58 0.000000
59 2020-04-15 sub_region ... 59 0.000000
60 2020-04-16 sub_region ... 60 0.000000
61 2020-04-17 sub_region ... 61 0.000000
62 2020-04-18 sub_region ... 62 0.000000
63 2020-04-19 sub_region ... 63 0.000000
64 2020-04-20 sub_region ... 64 0.000000
65 2020-04-21 sub_region ... 65 400.785828
66 2020-04-22 sub_region ... 66 401.776733
67 2020-04-23 sub_region ... 67 401.619812
68 2020-04-24 sub_region ... 68 403.111847
69 2020-04-25 sub_region ... 69 400.459259
70 2020-04-26 sub_region ... 70 402.953522
71 2020-04-27 sub_region ... 71 400.775208
72 2020-04-28 sub_region ... 72 400.355286
73 2020-04-29 sub_region ... 73 400.579224
74 2020-04-30 sub_region ... 74 401.702911
[25 rows x 28 columns]
INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/config.yaml INFO:wandb.run_manager:file/dir modified: /content/github_aistream-peelout_flow-forecast/wandb/run-20200528_183804-kvnf68wm/wandb-summary.json DEBUG:matplotlib.font_manager:findfont: Matching sans\-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=10.0. DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymReg.ttf) normal normal regular normal>) = 10.05 DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymBol.ttf) normal normal 700 normal>) = 10.335 DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono-Oblique.ttf) oblique normal 400 normal>) = 11.05 DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmex10' (cmex10.ttf) normal normal 400 normal>) = 10.05 DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBol.ttf) normal normal 700 normal>) = 10.335 DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-Bold.ttf) normal normal 700 normal>) = 0.33499999999999996 DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmsy10' (cmsy10.ttf) normal normal 400 normal>) = 10.05 DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans-BoldOblique.ttf) oblique normal 700 normal>) = 1.335 DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFiveSym' (STIXSizFiveSymReg.ttf) normal normal regular normal>) = 10.05 DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymBol.ttf) normal normal 700 normal>) = 10.335 DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans' (DejaVuSans.ttf) normal normal 400 normal>) = 0.05 DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXNonUnicode' (STIXNonUniBolIta.ttf) italic normal 700 normal>) = 11.335 DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneral.ttf) normal normal regular normal>) = 10.05 DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Display' (DejaVuSansDisplay.ttf) normal normal 400 normal>) = 10.05 DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXGeneral' (STIXGeneralBolIta.ttf) italic normal 700 normal>) = 11.335 DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmtt10' (cmtt10.ttf) normal normal 400 normal>) = 10.05 DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeFourSym' (STIXSizFourSymBol.ttf) normal normal 700 normal>) = 10.335 DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeOneSym' (STIXSizOneSymReg.ttf) normal normal regular normal>) = 10.05 DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeThreeSym' (STIXSizThreeSymReg.ttf) normal normal regular normal>) = 10.05 DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif' (DejaVuSerif-BoldItalic.ttf) italic normal 700 normal>) = 11.335 DEBUG:matplotlib.font_manager:findfont: score(<Font 'cmmi10' (cmmi10.ttf) normal normal 400 normal>) = 10.05 DEBUG:matplotlib.font_manager:findfont: score(<Font 'STIXSizeTwoSym' (STIXSizTwoSymReg.ttf) normal normal regular normal>) = 10.05 DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Serif Display' (DejaVuSerifDisplay.ttf) normal normal 400 normal>) = 10.05 DEBUG:matplotlib.font_manager:findfont: score(<Font 'DejaVu Sans Mono' (DejaVuSansMono.ttf) normal normal 400 normal>) = 10.05
Buffered data was truncated after reaching the output size limit.
df.tail()
| level | country | region | sub_region | date | lat | long | cases | deaths | recovered | active | tested | hospitalized | discharged | mobility_retail_recreation | mobility_grocery_pharmacy | mobility_parks | mobility_transit_stations | mobility_workplaces | mobility_residential | full_county | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 164400 | sub_region | United States | North Dakota | United States__North Dakota__Grand Forks County | 2020-03-29 | 47.9335 | -97.3975 | 0 | 0 | 0 | 0 | 400 | 0 | 0 | -46.0 | -20.0 | -65.0 | -38.0 | -28.0 | -2.0 | North Dakota_Grand Forks County |
| 164401 | sub_region | United States | Georgia | United States__Georgia__Wilcox County | 2020-04-13 | 31.9890 | -83.3945 | 24 | 0 | 0 | 0 | 0 | 0 | 0 | 0.0 | -6.0 | 0.0 | 0.0 | -30.0 | 0.0 | Georgia_Wilcox County |
| 164402 | sub_region | United States | Colorado | United States__Colorado__Routt County | 2020-03-23 | 40.4610 | -107.0345 | 3 | 0 | 0 | 0 | 0 | 0 | 0 | -71.0 | -34.0 | -44.0 | -65.0 | -51.0 | -2.0 | Colorado_Routt County |
| 164403 | sub_region | United States | California | United States__California__Mono County | 2020-04-08 | 38.0880 | -118.7410 | 20 | 1 | 0 | 0 | 103 | 0 | 0 | -100.0 | -68.0 | -63.0 | -69.0 | -57.0 | 0.0 | California_Mono County |
| 164404 | sub_region | United States | Florida | United States__Florida__St. Johns County | 2020-03-17 | 29.9375 | -81.4515 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | -1.0 | 25.0 | 22.0 | 9.0 | -30.0 | 9.0 | Florida_St. Johns County |
df[df['sub_region']==['United_States__California__Los_Angeles_County']]
--------------------------------------------------------------------------- ValueError Traceback (most recent call last) <ipython-input-37-0667b5d60a9c> in <module>() ----> 1 df[df['sub_region']==['United_States__California__Los_Angeles_County']] /usr/local/lib/python3.6/dist-packages/pandas/core/ops/common.py in new_method(self, other) 62 other = item_from_zerodim(other) 63 ---> 64 return method(self, other) 65 66 return new_method /usr/local/lib/python3.6/dist-packages/pandas/core/ops/__init__.py in wrapper(self, other) 527 rvalues = extract_array(other, extract_numpy=True) 528 --> 529 res_values = comparison_op(lvalues, rvalues, op) 530 531 return _construct_result(self, res_values, index=self.index, name=res_name) /usr/local/lib/python3.6/dist-packages/pandas/core/ops/array_ops.py in comparison_op(left, right, op) 232 # The ambiguous case is object-dtype. See GH#27803 233 if len(lvalues) != len(rvalues): --> 234 raise ValueError("Lengths must match to compare") 235 236 if should_extension_dispatch(lvalues, rvalues): ValueError: Lengths must match to compare
In order to make all the weights transferable we will need to redfine our Wandb sweep.
wandb_sweep_config_transfer = {
"name": "Default sweep",
"method": "grid",
"parameters": {
"batch_size": {
"values": [2, 3, 4, 5]
},
"lr":{
"values":[0.001, 0.002, 0.004, 0.01]
},
"forecast_history":{
"values":[10, 11, 12]
},
"out_seq_length":{
"values":[1, 2, 3, 4]
}
}
}
def get_most_recent_file(file_path):
list_of_files = glob.glob(file_path+"/*.pth") # * means all if need specific format then *.csv
if len(list_of_files) > 1:
latest_file = max(list_of_files, key=os.path.getctime)
return latest_file
return None
def format_corona_data(region_df:pd.DataFrame, region_name:str):
"""
Format data for a specific region into
a format that can be used with flow forecast.
"""
if region_name == 'county':
region_name = region_df['full_county'].iloc[0]
elif region_name=='state':
region_name = region_df['state'].iloc[0]
#else:
#region_name = region_df['country'].iloc[0]
region_df['month'] = pd.to_datetime(region_df['date']).map(lambda x: x.month)
d = pd.to_datetime(region_df['date'])
region_df['weekday'] = d.map(lambda x: x.weekday())
region_df['datetime'] = region_df.date
region_df.index = region_df.date
region_df = region_df.sort_index()
region_df = region_df.fillna(0)
region_df['new_cases'] = region_df['cases'].diff()
region_df.iloc[0]['new_cases'] = 0
region_df= region_df.fillna(0)
region_df.to_csv(region_name+".csv")
print(region_df.head(9))
return region_df, len(region_df), region_name+".csv"
def run_full_geo_code(df_list, start_index, end_index, use_transfer=True):
for i in range(start_index, end_index):
file_path, len_df, file_path_name = format_corona_data(df_list[i], 'county')
latest_file = get_most_recent_file("model_save")
sweep_full = wandb.sweep(wandb_sweep_config_transfer, project="covid_forecast", entity='covid')
if use_transfer and len(os.listdir("model_save"))>1:
print("using transfer")
wandb.agent(sweep_full, lambda:train_function("PyTorch", make_config_file(file_path_name, len_df, weight_path=latest_file)))
else:
wandb.agent(sweep_full, lambda:train_function("PyTorch", make_config_file(file_path_name, len_df)))
The purpose of this experiment is mainly to tune the MultiHead weights